%%
}

 

 

 

RATIO

Ay
 

 

New York
State Callege of Agriculture

At Cornell University
Sthaca, N. Y.

 

Library
  
 
 
   
   
    
 
   
    

ou 18, No.9 PSYCHOLOGICAL REVIEW PUBLICATIONS September, 1921

sychological Bulletin

EDITED BY

SHEPHERD I. FRANZ, Govt. Hosp. ror INSANE
SAMUEL W. FERNBERGER, UNIV. oF PENNSYLVANIA (Assistant Editor)

OWARD C. WARREN, PRINCETON UNIVERSITY (Review)
JOHN B. WATSON, New York (/. of Exp, Psych.)
JAMES R. ANGELL, YALE University, (Monographs) aun
MADISON BENTLEY, University or ILLINOIS (/#dex)

WITH THE CO-OPERATION OF

. T. BALDWIN, University of Iowa; E. B. HOLT, Brunswick, Mg,; W. S. HUNTER
Hf rstversiry oF Kansas; J. H. LEUBA, Bryn Mawr CoLteceE; MAX MEYER, UNI-
AA eestry oF Missourr; R. M. OGDEN, CornELL University; W. D. SCOTT, NortTH-
i

7ESTERN UNiversiTy; G. M. WHIPPLE, UNiversity oF ILLINoIs; R. S. WOODWORTH
om UNIVERSITY.

CONTENTS

eneral Reviews :

Personality and Character: G. W. ALiport, 441. Mental Work: ES.

 

ROBINSON, 456. Recent Literature on the Psychology of the Musician :
M. ScHoen, 483. Zhe Psychology of Language: E. A. Esper, 490.

ectal Reviews :
Watt s Foundations of Music: R. M. OGDEN, 497.

PUBLISHED MONTHLY BY THE

PSYCHOLOGICAL REVIEW COMPANY
41 NORTH QUEEN ST., LANCASTER, PA.,
AND PRINCETON, N. J.

Aacgnts: G. BE, STECHERT & CO., Lonvon (2 Star Yard, Carey St., W. C.); Pamus (16, rue de Condé)

Entered aa second-class matter January 31, 1904, at the post-office at Lancaster, Pa., under the
Act of Congress of March 3, 1879
Psychological Review Publications
S614 97 :

EDITED BY

HOWARD C. WARREN, PRINCETON UNIVERSITY (Review)
JOHN B. WATSON, 244 Maptson AvE., New York (J. of Exp. Psych.)
JAMES R. ANGELL, Yale UNIVERSITY (Monographs)
SHEPHERD I. FRANZ, Govr. Hosp. For INSANE (Bulletin)
MADISON BENTLEY, University or ILiinots (Index)

WITH THE CO-OPERATION OF
Many DIsTINGUISHED PsyCHOLOGISTS

PSYCHOLOGICAL REVIEW

containing original contributions only, appears bimonthly, January, March,
May, July, September, and November, the six numbers comprising a volume of
about 480 pages.
PSYCHOLOGICAL BULLETIN

containing abstracts, critical reviews of books and articles, psychological news
and notes, university notices, and announcements, appears monthly, the annual
volume comprising about 720 pages. Special issues of the BULLETIN consist
of general reviews of recent work in some department of psychology.

JOURNAL OF EXPERIMENTAL PSYCHOLOGY

containing original contributions of an experimental character, appears bi-
monthly, February, April, June, August, October, and December, the six num-
bers comprising a volume of about 480 pages.

PSYCHOLOGICAL INDEX

is a compendious bibliography of books, monographs, and articles upon psycho-
logical and cognate topics that have appeared during the year. The INDEX is
issued annually in May, and may be subscribed for in connection with the peri-
odicals above, or purchased separately.

PSYCHOLOGICAL MONOGRAPHS

consist of longer researches or treatises or collections of laboratory studies which

it is important to publish promptly and as units. The price of single numbers

varies according to their size. The Monocrapus appear at irregular intervals

and are gathered into volumes of about 500 pages.

Philosophical Monographs: a series of treatises more philosophical in character.
Library of Genetic Science and Philosophy: a series of bound volumes.

ANNUAL SUBSCRIPTION RATES

Review: $4.25 (Foreign, $4.50). Review and Bulletin: $9.00 (Forei oO

Jone £425 Coe oe on pe $8.25 Green ee

etin: $5.00 (Foreign, $5.25). Journal and Bulletin: $o0, i

Any one of above with Index: $1.25 additional. Be A EEE Eee
Any two of above with Index: $1.00 additional.

Review, Bulletin, and Journal: $12.50 (Foreign, $13.25).

Review, Bulletin, Journal, and Index: $13.50 (Foreign, $14.25).

Current Numbers: Review or Journal, 80c; Bulletin, 50c; Index, $1.50
Psychological Monographs: $5.50 per volume (Foreign, $5.80). ;
Current Issues: prices vary according to size.

Subscriptions, orders, and business communications may be sent direct to

PSYCHOLOGICAL REVIEW COMPANY

41 N. QUEEN ST., LANCASTER, PA., or PRINCET
Foreicn AGEnts: G. E. STECHERT & CO., London (2 Star Yond Ge St
W.C.); Paris (16, rue de Condé) (P. 2A

IT
CARNEGIE
Institute of Technology

APPLIED PSYCHOLOGY

Industrial Psychology Statistical Methods
Educational Psychology Personnel Administration

Carnegie Institute of Technology makes pro-
vision for graduate students who come already
well grounded in Systematic, Experimental and
Physiological Psychology to specialize in Applied
Psychology.

Applications to industry, to business, and to
vocational education are stressed. Two years of
specialized study and research prepares for a
university instructorship in Applied Psychology,
or for an assistantship, either research or execu-
tive, in a large business or industrial concern.

Annual tuition, $150.

THE STAFF, 1921-1922

W. V. BINGHAM W. W. CHARTERS

C. S. Yoakum E. K. STRONG, JR.

L. L. THURSTONE Marion A. BILLs

W. R. SKILLEN H. G. Kenacy
ELIZABETH DYER PAULINE A. HUMPHREYS
D. R. CRAIG M. J. REAM

C. F. HANSEN Max SCHOEN

B. W. ROBINSON C. C. CRAWFORD

Address: DirEcToR W. V. BINGHAM

Carnegie Institute of Technology
Pittsburgh, Pa.

(P, B.)
Che Johns Hopkins University

DEPARTMENTS OF PHILOSOPHY, PSYCHOLOGY
AND EDUCATION

GRADUATE COURSES, 1920-1921

PHILOSOPHY. Students maytake for their principal subject general
systematic philosophy, logic and scientific methodology, ethics, or the
history of philosophical ideas, both in the technical systems and in
literature and the sciences. An important part of the student’s work
will each year consist in imdividual study on selected topics (systematic
reading, methodical analysis of problems, frequent written papers) ' nder
the supervision of one of the professors of the department. Other
courses will be announced later.

PSYCHOLOGY. Provision is made for Research and for Graduate
Instruction. Courses are offered in Experimental Psychology by Knight
Dunlap; in Child Psychology by Buford Johnson and in The Psychology
of Music by Otto Ortman. Opportunities for research in Experimental
Psychology, Child Psychology and Social Psychology are offered.

EDUCATION, Opportunities for instruction and investigation are
offered to meet the needs of students whose interest leads them to
specialize in the fields of Educational Administration, Educational Psy-
chology, Experimental Education, or Educational Theory and Practice.
Courses are offered in 1920-1921 as follows: Educational Administration
and Experimental Education, Edward F. Buchner ; Modern Educational.
Theory, Florence E. Bamberger ; Secondary Education, David E. Weglein;
Supervision of Elementary Education, Florence E. Bamberger.

For Philosophy, Professor A. 0. LOVEJOY
For Psychology, Professor KNIGHT DUNLAP
For Education, Professor EDWARD F. BUCHNER

(P. B.)

iv
Vol. 18, No. 9 September, 1921
THE

PSYCHOLOGICAL BULLETIN

PERSONALITY AND CHARACTER
BY GORDON W. ALLPORT

Harvard University

Psychologists of the past decade have been much occupied with
the application of their science to practical problems of life, partic-
ularly to problems of vocation and conduct. The social adjust-
ments of the individual represented in such problems are receiving
marked attention not only from clinical and consulting psycholo-
gists, but from experimental and research psychologists as well.
The enticing topic of speculation, human nature, has been rephrased
for scientific study as a problem of individual nature. Onevery hand
we see that interest is becoming focused upon the study of man
with the hope that by this method the problems of the mechanisms,
motives, and modifications of behavior peculiar to the individual
may at length be solved. This tendency is manifested by the rapid
ncrease in the number of studies of human personality and character.

This paper is prepared with the hope that it may offer a useful
summary of experimental and clinical researches, together with a
review of the best of the speculative work, and so assist the reader
to a clarified notion of the present status of psychological investi-
gations concerning personality and character. For convenience the
contributions to be considered will be arranged under three headings:
(1) the psychological nature of character and personality (the
problem of definition); (2) the composition of personality (the
classification of traits); and (3) the measurement of traits by testing
devices and rating scales.

J. Tue Psycuotocica, Nature oF PERSONALITY
AND CHARACTER

In this field of study we may expect richer results from experi-
mental and descriptive work than in the direction of definition.

441
442 GORDON W. ALLPORT

Indeed, the problem of definition is almost hopelessly complicated
by prevalent popular uses of terms. Recently, however, there have
been noteworthy efforts made to delimit carefully the concepts of
personality, character, and temperament as to their psychological
significance.

Paton (44) considers personality to be the dynamic or functional
aspect of individuality—“the progressive synthesis of the mosaic
of experiences.” Individuality which is mere organismal unity
expresses only structural differentia between members of a species;
whereas personality is in all cases the peculiar function of the indi-
vidual. This view is similar to Kempf’s (31) who considers per-
sonality to be the habitual modes of adjustment which the organism
effects between its own egocentric drives and the exigencies of the
environment. Between the autonomic system and the exactions of
the physical and social environments we find the intermediation of
the central nervous system, and the customary adjustments which
this system makes between the egocentric drives of the organism
and the requirements of the milieu constitute the personality of
the organism. Such a standpoint serves to call attention to the
biological fact that the primitive nervous system (the autonomic)
is of immense importance in determining personality. The view as
stated compels us to ascribe personality to all living things, unless
we limit the concept of “‘habits of adjustment” to the social en-
vironment, leaving out of account the numerous adaptive responses
made to non-social surroundings. Limiting the concept in this
way would be to consider personality as developing pari passu with
the number and complexity of social relationships.

There are many rag-bag theories of personality, into which are
dumped “‘all biological innate dispositions, impulses, tendencies,
appetites and instincts, and all acquired dispositions” (49). Such
a conception partakes of chaos, for it does not give unity or definite-
ness of direction toour study. Itis not difficult for the self observer
to recognize that his own personality is something more homo-
geneous and dynamic than such a definition would suggest. Nor
will the observer be content with the static conception that “per-
sonality is the totality of all our previous experiences and ideas.”
Hartenberg (22) introduces an organizing motive into the rag-bag.
He considers personality to be the peculiar organic condition of
the individual, with his individual attributes of sensory and motor
behavior, capped and organized by a dominant sentiment or motive.

Myerson’s book (40) seeks to analyze the foundations of per-
sonality, but does not attempt to distinguish it from character or
PERSONALITY AND CHARACTER 443

mind, since these are all alike viewed as functions of the organic
activities of the body. It is to be regretted that the author de-
spaired of defining his field a little more accurately, for there is
much ambiguity entailed by confusing three such totally different
problems.

The distinction between personality and character is an impor-
tant one for psychologists to observe. Webb (64) defines character
as the sum of all personal qualities which are not distinctly intel-
lectual. In this regard he agrees with Fernald (15) who suggests
that the formula for personality is intelligence plus character. Mc-
Dougall (34) considers character to be the organization of senti-
ments, which are innate, into a system or hierarchy within the
individual. According to this view character is not as unitary a
function as Fernald would have us believe, nor can it be thought
to be measureable. Gosline (20) considers character, along with
intelligence and temperament, to be inborn, though susceptible to
alteration by the environment. On this assumption character be-
comes a behavior function comparable with intelligence as a field
for psychological investigation. In contrast to these views we
find that of Watson (62). Character is defined by this author as
the personality evaluated according to prevailing standards of
conduct. Those psychologists who accept Watson’s view have no
right, strictly speaking, to include character study in the province
of psychology; it belongs rather to social ethics.

Watson considers personality to be the sum total of an indi-
vidual’s responses to his environment, that is to say, his “reaction
mass”? asa whole. Kantor (30) distinguishes two types of behavior
within the personal adjustment habits, (a) actions which are direct
visible signs of the individual’s nature, and (b) more permanent
action elements, such as dispositions to action which may be con-
sidered as potential behavior. Such views as these, placing em-
phasis upon adjustment habits, have been at the base of the numer-
ous attempts recently made to measure traits and to study the
behavior patterns which constitute the personality. (Part III of
this review).

II. Tue Composition oF PERSONALITY

What traits do psychological writers recognize as the elementary
constituents of personality? It has long been considered that traits
are mental faculties, which might be studied by reference to the
characteristic configuration of head or hand. It is not within the
444 GORDON W. ALLPORT

scope of this paper to delve into these “slums of psychology”:
the reader who is interested will find profitable a review by Jastrow
(27) of the antecedents of character study.

Psychologists are working steadily toward a schedule of human
traits which will express in the most economical fashion a true
analysis of personality. Wells (66) has gathered together the classi-
fications of Heymans and Wiersma (23), Cattell (8), Hoch-Amsden
(24), and Davenport (10), and has presented a synthesized schema
of traits. The result is a list of fourteen types of adjustments
which are fundamental to every personality. These essential modes
of behavior are then divided into ninety-four traits which submit
to quantitative study according to a rating scale. Partridge (43)
and Yerkes and La Rue (71) further multiply the number of qual-
ities, and so invite upon their schemes the criticism of superficiality
and overlapping. In fact, a notable weakness at the present time
is the tendency to make a hasty and inadequate selection of traits.

Spaulding (55) represents one of the many recent attempts to
find a practical schedule of traits for application to a special problem
—in this case, the problem of delinquents. The study presents
one hundred thirty-four qualities arranged under eight principal
divisions of personality (intelligence, emotionality, and the like).
Porteus (46, 47) has made a similar attempt to discover the qualities
which are essential for the social adaptation of defectives. Business
leaders, social workers, educators, moral directors, are continually
publishing personal analysis charts. The same criticism can be
made of one and all: the traits chosen are not psychologically
basic; they represent complexes of habits (neatness, tact, and the
like) rather than truly fundamental aspects of personality. Wood-
worth (70) presents a plan of four subdivisions, physique, chemique,
instincts, and intelligence—more suggestive than the four humors
upon which most of the French studies of the caractéres are based
—but still very schematic.

If we turn our attention from the classifications in toto to some
of the special traits which are frequently advocated to be elementary
constituents of personality, we may then be able to see the problem
more clearly. It is dangerous to oversimplify the problem of
classification, and in the discussion which follows it must be remem-
bered that by no means all of the alleged components of personality
are included. The traits chosen are merely those upon which a
large number of students of personality are focusing attention at
the present time.
PERSONALITY AND CHARACTER 445

Intelligence —No scheme of personality denies a signal position
to the intellectual capacity. Terman (57) and Webb (64) find
that high intelligence correlates in general with all strong qualities
of character. However intelligence may be defined, there can be
no doubt that the mental adaptibility for which it stands is one of
the major factors in the development of personality.

Temperament.—The emotional characteristics of an individual
are likewise accorded universal recognition. ‘There is, however,
little agreement as to what temperament is. The four-humor
doctrine has haunted the house of psychology for twenty-five
centuries. The advent of endocrinology, as Paton (44) points out,
has done much to supply a reinterpretation of the theory, and as
the work progresses we may find that the anticipations of Hippoc-
rates and Galen were not so far afield. There has been a steady
increase of emphasis upon the réle which the ductless glands play
in determining the emotional life. Berman (4) has recently pub-
lished a mechanistic interpretation of personality almost entirely in
terms of glandularinfluences. He speaks of the thyroid personality,
the pituitary personality, and the gonado-centric personality, ac-
cording as these autocoid influences seem to predominate in the
individual!

Davenport (9) suggests that qualities of emotionality, ¢.g., a
“hot temper,” may be inherited according to the Mendelian prin-
ciple. Heymans and Wiersma (23) are quite certain that their
findings based upon a Massenuntersuchung show evidence of the
positive inheritance of traits of character and temperament. Gates
(19) has reviewed the current beliefs concerning the inheritance of
mental traits in this journal.

Volitional Qualities—Downey (12) has called attention to the
significance in personality of the factors of activity and control.
Myerson (40) employs this conception of the “kinetic” nature to
describe personality types which seem to be characterized by a
certain speed or intensity of response. Prominent among these
“volitional”? types we find the ascendant personality and the sub-
missive personality, characterized respectively by habitual aggres-
sion or passivity. The analogy of the active and passive voice in
these types was first suggested by Southard (54). Allport (2) has
applied this conception to the problem of salesmanship, showing
that timidity in the selling personality is essentially passivity, and
this is to be overcome only by a deliberate building up of one’s
ascendant qualities. Ascendance is the essential trait of leader-
446 GORDON W. ALLPORT

ship. Gowin (21) has found it to be correlated with physical attri-
butes of size, weight and personal appearance.

There are other characteristics of the personality which are of
great importance, but which have not been adequately studied.
Such, for example, is the esthetic nature. Again, we have the
expansive personality as contrasted to the recluse. There has been
no study of these traits to the present writer’s knowledge, excepting
his own research (3) in the Harvard Psychological Laboratory.
(See Part III of this paper.)

Reference should be made to two or three notable contributions
from psychiatrists and Freudian psychologists. White (69) con-
siders conflict in the psychoanalytic sense to be the cardinal problem
of character formation. The well-balanced personality is the one
which effects a successful “resolution” of all conflicts. Adler (1)
has advanced the theory of Minderwertigkeit, according to which
each individual who finds himself handicapped in life’s struggles
because of physical or mental defects will tend to compensate for
this consciousness of personal inferiority. The inherent desire of
a person to be superior and powerful leads him to make many
peculiar attempts at adjustment, to preserve at all costs his “feeling
of masculinity,’ and the characteristic habits which result are true
traits of personality. The biologist, Le Dantec (32) likewise holds
egoism to be the determinant of all actions, and the individual’s
personality merely represents the “deformations” which la vie en
commun compels him to undergo. Adler’s theory has been inter-
preted and popularized by Tanner (56), Myerson (40) and others,
until Compensation has become one of the psychological headlines
of the day.

Jung (28) has drawn attention to two important types of
adjustment. The extravert finds his real life one of keen partici-
pation in his environment, the introvert on the other hand living
much within the world of self-consciousness and fantasy. These
types, those whose adjustment is realistic, and those whose adjust-
ment is autistic are considered in detail by Wells (65). Day-
dreaming he holds to be one of the characteristic attempts at inner
adjustment when outer adjustment fails.

Rosanoff (50) gives an excellent account of psychopathic per-
sonality types. ‘There are individuals whose personality might be
characterized as (1) antisocial, (2) cyclothymic, consisting of manic
make-up, depressive make-up, irascible make-up, or emotional in-
PERSONALITY AND CHARACTER 447

stability, (3) autistic, i.e, predominantly introverted, and (4) the
epileptic personality, with its periodic alterations of mood and
character. Normal personalities are considered by Rosanoff to
differ from the psychopathic only slightly as to quality but markedly
as to quantity. The schizophrenics and epileptics, it is true,
evidence brain atrophy, but in regard to most traits there appears to
be a perfect continuity of type from the most normal individual
to the totally insane. Southard (53), has made a suggestive innova-
tion by the application of grammatical categories to personality
study. Certain features of the personality, he says, may be studied
with profit by describing them in the terms of syntax: person,
number, mood, voice, and tense.

III. Tue Measurement or PERSONALITY BY RaTING
‘AND TESTING

In the field of personality measurement we find an encouraging
amount of meritorious research. In all probability, as has been
the case with the study of intelligence, we shall be able to give
reliable quantitative results before we understand the precise nature
of that with which we are dealing. Even before the days of Freud-
ian-endocrine-psychiatric enlightenment as to the probable nature
of personality, Francis Galton definitely advocated psychometric
tests for traits of character. He suggested, for example, that it
might be profitable to construct a delicate apparatus for recording
individual peculiarities of manner in sitting upon chairs. Such a
device would most certainly yield important information concerning
the sitter!

1. The Rating of Traits—Norsworthy (42) seems to be a
pioneer in the study of the validity of judgments upon character.
Her conclusions, substantiated in the main by subsequent investi-
gations, are (1) that the reliability of rating varies with the traits
under consideration. Raters agree more closely upon such qualities
as popularity, conceit, or leadership than upon emotionality, hon-
esty, or tact. (2) It seems also that some individuals are easier
to rate and their rank more readily determined by the raters than
other individuals. We have thus a suggestion of the open person-
ality versus the enigmatic. Cattell (8) decides that traits upon
which the judges closely agree represent the individual’s reactions
to objective things, whereas the traits upon which they disagree
most represent the individual’s reactions toward other people.
448 GORDON W. ALLPORT

A study made by Thorndike (59) sounds a warning which all
dealers in rating scales may well heed. It was found in his study
that those giving ratings were unable to analyze out the different
aspects of an individual’s nature and rate them independently of
other aspects. The ratings are affected by a marked tendency to
think of the person in general as good or rather inferior, and so
color the judgments of the qualities by a general feeling. Thus
a “halo” is provided the individual who is being rated. Moore
(36) finds evidence of this halo as an influence upon the rating of
college students by members of the faculty. This same source of
error helps to account for the conclusions reached by Rugg (52)
that the rating scale used in the U. S. Army (45) probably did not
locate an officer within his “fifth” of the entire scale. The scale
is reprinted in this journal, and a discussion of the theory and appli-
cations is presented in the same number by Terman (58). Rugg
maintains that to render rating of human character practicable,
the following conditions should be fulfilled: (1) The final rating
must be the average of three independent ratings, each one made
on a scale as objectified as the man-to-man-comparison type of
scale. (2) The scales on which ratings are made must be compar-
able and equivalent, having been made in conference under the
instruction of one skilled in rating scale work. (3) The raters must
be so throughly acquainted with the person rated that they are
competent to make the judgment. The question of reliability
has not deterred schools and industries from using rating scales,
and whichever way we turn we meet with these schemes (41, 60,
63). Wells (66) and Spaulding (55) endeavor to eliminate the
difficulties of scoring by the employment of a simple + and —
designation opposite each trait in their scales, indicating superiority
or deficiency of the attribute as compared to the average or normal
person.

Self-rating is fraught with perils. Hollingworth (25) finds that
the individual generally tends to overestimate his possession of
qualities which are socially valuable and to underestimate in the
case of those which are socially undesirable. Thus, no person
would be content to ascribe to himself anything but a superior
possession of refinement or humor. Hollingworth found it to be
a rule that the individual’s deviation in self-ranking from the place
accorded him by averaging the ratings of his associates is noticeably
greater than the average deviation among the judgments of the
associates themselves. ‘The possibility should be noted that the
PERSONALITY AND CHARACTER 449

extent to which the self-rating deviates from the average rating by
associates may be used as an index of the insight of the person
studied. The lower the average of his deviations from the rating
accorded him on various traits, the higher his insight. In cases
where ratings are made on desirable or undesirable qualities a Self-
evaluation or Conceit Index might be computed in a similar fashion.

Notwithstanding the dangers and difficulties encountered in
devising and employing rating scales, we are forced to recognize
this method as the only available objective criterion of personality.
The sources of error must gradually be overcome by the improve-
ment in the technique of rating. The methods of measuring ob-
jective manifestations of personality which we shall now turn
attention to, are as a rule more unreliable than ratings by associates.
The only method of judging the value of a particular test is by its
correlation with ratings. Since both factors are generally so vari-
able, we cannot expect more than very modest correlations from
even the most refined test methods. ‘There is cause for optimism,
however, in the keen and original methods which are being devised
for obtaining objective proof of the presence of a given trait and
for approximate measure of its magnitude.

2. The Testing of Traits—The development of the age and
point scale standards for the measurement of intelligence has made
clear to psychologists the imperative need for reliable methods of
studying other aspects of personality. To supplement the 1.Q.
we need reliable measures of emotional, volitional and social qual-
ities. Brubacher (5) suggests the concept of a Personality Quotient,
and applies his idea to the qualities requisite to the teaching pro-
fession. It must be remembered, however, that differences of
personality are of a qualitative as well as of a quantitative sort.
This difficulty stands in the way of the development of a measure-
ment scale based on the correlation between tests and familiar
objective criteria such as those of intelligence. Methods which
have been employed aim for the most part at a sampling of cus-
tomary reactions, and the results obtained contribute rather more
to a descriptive treatment of personality than to a quantitative
analysis.

The Word Association Method.—Jung (29) brought to our
attention the possibility of studying egocentric and objective types
by their characteristic responses in the word reaction experiment.
Wells (68) has published a bibliography of German studies which
have dealt with the relation between the association method and
450 GORDON W. ALLPORT

the study of temperament. The same author (67) has made a
study of the egocentric and objective types, with the discovery
that the former tends to give the more unusual responses as meas-
ured by standardized frequency tables. Complex types of intro-
version are studied in reference to diagnostic criteria by Hull-
Lugoff (26). In this study it was found that the most reliable
single indicator of the tapping of a complex is the tendency of the
subject to repeat the stimulus word. Other specially valid criteria
are a prolongation of reaction time (over 13/5 seconds), inability
to make any response whatever, and extremely short reaction times.
The findings of Dooley (11) show that long reaction times in general
seem to go with the egocentric type, and that the habitually quick
reaction is a sign of objectivity in the reagent. In an experiment
with directed egocentric reactions, Washburn and others (61) as-
sumed that subjects who can make a required personal association
with the least delay are egocentric. The present writer in a recent
experiment has not found this assumption to be justified.

Moore (35) uses the association method to test the strength of
instincts. Stimulus words are arranged into constellations accord-
ing to McDougall’s schedule of instincts. The subject is asked to
make a personal association which is scored on the basis of indicating
strength or weakness in the possession of the instinct in question.
Morgan and others (38) presented stimulus words to subjects for
five successive mornings, and secured the response from them as to
whether the words were pleasant or unpleasant in tone. An index
of optimism or pessimism was thus computed which showed a
certain correlation with the ratings on this trait by associates.
Not unlike this method is the study of personality through an
analysis of the content of images by Martin (33). Mental and
physical peculiarities and preferences are considered to be subject
to profitable study by this method.

Measurement of Motor Expression.—The relation of motor
impulses to psychic patterns has suggested several experimental
attacks. Much work has been based upon the theory that the
petites gestes involved in handwriting are movements as expressive
of personality traits as are gross bodily movements. The work of
Klages, Binet, Crépieux-Jamin and other graphologists has been
summarized by Downey (14). This author herself made a careful
study of handwriting and concludes that it bears certain definable
relations to character. Some of the types to be studied by this
method are the Explosive, Obstructed, Sensory, Motor, Hyper-
PERSONALITY AND CHARACTER 451

kinetic, and Hypokinetic. A careful refinement of technique in
the study of handwriting is necessary, for gross features are often
misleading. “The broad classification of outgoing movements and
movements of withdrawal as characterizing respectively attitudes
of aggression and of defense cannot sustain too great a weight.”
Downey (12, 13) has designed and partially standardized a Test
for Will-Temperament. Numerous factors such as speed, size,
disguise and control in handwriting are measured carefully and
weighted. The scores for the various portions of the test may be
plotted into a Will-Profile. Bryant (6) in a study of delinquent
boys has shown that by Downey’s method it is possible to secure
an approximate measurement of the volitional qualities which make
for social or antisocial conduct. Ruch (51) however, finds that the
actual correlations between the scores on the Will-Temperament
“est and student and faculty ratings are low. The Carnegie
Institute of Technology (7) has adapted the Downey test for
practical use in business, one of the many indications of the current
‘enthusiasm for applying psychological methods.

“Will power” is studied by Fernald (16) by means of an appa-
ratus for measuring the ability of the subject to stand upon his
toes for a long period of time. The problem of the shifty eye in
relation to the lack of aggressiveness has been studied by Moore
(37), with the conclusion that, taken together with certain other
diagnostic signs, the number of eye movements noted while the
subject performs a given mental problem is indicative of his ascen-
dant or submissive attitude in general.

Represented Situations.—One of the methods of testing char-
acter dispositions definitely advocated by Galton is that of repre-
senting in miniature certain problems of actual life, and of observing
the individual’s adjustment to these situations. Freyd (18) em-
ploys representative problems in news getting as one of a series of
nine tests for journalistic aptitude. Fernald (16) presents to a
delinquent subject ten questions concerning ethical interpretations,
and finds that the ability to make a discrimination according to
the conventional code in the case of these samples is an approximate
measure of the subject’s ethical adjustment as a whole. Allport (3)
has used this method as a test of ascendance-submission, and finds
that the scores on sample problems representing the individual’s
behavior in situations which would require his taking an active
or a passive réle, correlate slightly with personality ratings by
associates on this trait (r = .40). Myerson (39) suggests limiting
452 GORDON W. ALLPORT

the number of possible answers to a miniature problem, say, to
four. This “multiple choice” device facilitates scoring, and sug-
gests at once which of four possible types the reagent belongs to.
Ethical discrimination and humor are the traits which Myerson
has most extensively studied by this method. Pressey (48) has
published a test for emotional spread. Lists of mixed words, strong
and weak, are presented to the subject who is required to cancel
out all which have unpleasant feeling tones associated. The “affec-
tive spread,” i.¢., the range of situations to which the individual
is wont to react in an emotional manner, is a distinctly serviceable
concept. Pressey’s study stands out as practically the only attempt
so far made to measure the complex emotional life of the individual
by a simple testing device.

Questionnaires.—The questionnaire method of studying person-
ality has, with all its limitations, definite value in checking on the
results of tests, and for suggesting traits which otherwise might
escape notice. One of the revelations, for example, which can be
made by the answers to a questionnaire concerns the factor of
compensation. This important mechanism defies direct experimen-
tation, and can be revealed only by a historical study of the person
such as a carefully devised questionnaire affords. Perhaps the
completest lists of questions are Watson’s (62) and Woodworth’s,
the latter to be found in Franz’s handbook (17). Both of these
questionnaires have the advantage of being couched in ‘‘behavior-
istic” language, and the latter lends itself to simple “‘yes” and
“no” responses. However highly this method may be developed,
as an instrument for accurate study of the personality it will always
face the weakness of permitting falsification or rationalization on
the part of the subject.

REFERENCES

1. Apter, A. The Study of Organic Inferiority and its Psychical Compensation.
(Transl. Jeliffe). Nerv. & Ment. Disease Monograph, 24, 1917.

2. Autrort, F.H. Timidity and the Selling Personality. The Eastern Underwriter,
1920, 21, 15-17.

3. Auuport, F. H. & G. W. Personality Traits: Their Classification and Measure-
ment. J. of Abnor. Psychol. &% Soc. Psychol., 1921, 16, 6-40.

4. sei L. The Glands Regulating Personality. New York: Macmillan, 1921.

Pp. 291.

5. Brupacuer, A. R. The Teaching Personality Quotient. J. of Appl. Psychol.,
1921, 5, 39-45.

6. Bryant, E. K. The Will Profile of Delinquent Boys. J. of Deling., 1921, 6,
294-309.

7. Carnegie Inst. Technol. Will-Temperament Test. Bureau of Personnel Research,
1920, Test No. 9.
10.
II.

12.

13.

14.

15.

16.

17.

18.

19.

20.

21.

22.
23.

24.

25.

26.

27.

28.

29.
30.

31.

32.
33-

34.

PERSONALITY AND CHARACTER 453

. CaTTELL, J. American Men of Science. New York: Science Press, 1910. Pp.

596.

» Davenrort, C. B. Heredity in Constitutional Mental Disorders. Psycuot.

BULL., 1920, 17, 300-312.

Davenrort, C. B. The Trait Book. Eugenics Record OF. Buil., No. 6, 1912.

Dootey, L. Correlation of Normal Complexes. Amer. J. of Psychol., 1916, 27,
III-139.

Downey, J. The Will Profile: A Tentative Scale for Measurement of the Voli-
tional Pattern. U. of Wyom. Bull., 1919, 16, 1-40.

Downey, J. Some Volitional Patterns Revealed by the Will Profile. J. of
Exper. Psychol., 1920, 3, 281-301.

Downey, J. Graphology and the Psychology of Handwriting. Baltimore: War-
wick & York, 1919. Pp. 138.

Fernaup, G.G. Character as an Integral Mentality Function. Ment. Hygiene,
1918, 2, 448-462.

Fernaup, G. G. The Delinquent Class: Differentiating Tests. Amer. J. of
Insanity, 1912, 68, 523-594.

Franz, 8.1. Handbook of Mental Examination Methods. (2d Edit.) New York:
Macmillan, 1919. Pp. 193.

Freyp, M. Test Series for Journalistic Aptitude. J. of Appl. Psychol., 1921, 5,
46-56.

Gates, A. I. The Inheritance of Mental Traits. Psycuou. But., 1921, 18,
358-365.

Gosurne, H. I. Personality from the Introspective Viewpoint. J. of Abnor.
Psychol., 1920, 15, 36-44.

Gowin, E. B. The Executive and his Control of Men. New York: Macmillan,
1915. Pp. 349.

Harrensere, P. Physionomie et Caractere. Paris: Alcan, 1911.

Hermans & Wiersma. Beitrage zur Speziellen Psychologie auf Grund einer
Massenuntersuchung. Zsch. f. Psychol., 1906, 42, 81-127; 253-301.

Hocu-Amspen. A Guide to the Description and Study of the Personality. State
Hosp. Bull., New York. 1913.

Hotunewortn, H. L. Vocational Psychology. New York: Appleton, 1916.
Pp. 302.

Huur-Lucorr. Complex Signs in Diagnostic Free Association. J. of Exper.
Psychol., 1921, 24, 414-425.

Jastrow, J. The Antecedents of the Study of Character and Temperament.
Pop. Sct. Mo., 1915, 87.

June, C. Analytical Psychology. London: Bailliere, Tindall & Cox, 1916,
Pp. 377-

June, C. Studies in Word Association. London: Heinemann, 1918. Pp. 560.

Kantor, J. B. Human Personality and its Pathology. J. of Phil., Psychol., &
Sct. Meth., 1919, 16, 236-246.

Kemer, E. J. The Autonomic Functions and the Personality. Washington:
Nerv. §S Ment. Monog. Series, No. 28, 1918. Pp. 1-151.

Le Dantec, F. L’Egoisme. Paris: Flammarion, 1918. Pp. 324.

Maaztin, L. J. Personality as Revealed by the Content of Images. Science, 1917,
45, 393-399.

McDovcatt, W. Social Psychology. Boston: Luce, 1921. (14th Edit.) Pp.
418.
454 GORDON W. ALLPORT

35+
36.
37-
38.
39-
40.

41.
42.

43-
44.
45-
46.
47-
48.

49.
50.

51.

52.
53-
54:
55-

56.
57-

58.
59:

62.

Moors, H. I. A Method of Testing the Strength of Instincts. Amer. J. of
Psychol., 1916, 27, 227-233.

Moors, H. I. Three Types of Psychological Rating in Use with Freshmen
at Dartmouth. School & Soc., 1921, 13, 418-420.

Moors, H. I. The Measurement of Aggressiveness. J. of Appl. Psychol., 1921,

» 97-118.

Py E. Test of Moods and Temperaments by Recall of Emotionally Toned
Experiences. Amer. J. of Psychol., 1919, 30, 302-304.

Myerson, A. Multiple Choice Method of Studying Personality. Arch. f. Neurol.
und Psychopathol., 1919, 459-470.

Myerson, A. The Foundations of Personality. Boston: Little, Brown, 1921.
Pp. 400.

eae Charts. Chevy Chase, D. C.: National Inst. for Moral Instruction, 1920.

Norswortuy, N. The Validity of Judgments of Character. Essays in Honor of
James, 1910. (2d Edit.) Pp. 542-552.

Partripce,G.E. An Outline of Individual Study. New York: Sturgis & Walton,
I910. Pp. 240.

Paton, 8. Human Behavior. New York: Scribners, 1921. Pp. 454.

Personnel System, U.S. Army. Rating Scale. 1919, 1, 559-581.

Porteus, 8. D. A Study of Personality of Defectives with a Social Rating Scale.
Public Train. School, 1920, 23, 1-22.

Portus, 8. D. Personality in Relation to Social Mal-adjustment. Training
Sch. Bull., 1921, 18, 81-90.

Pressey, S.L.& L.W. First Revision of a Group Scale Designed for investigating
the Emotions, with Tentative Norms. J. of Appl. Psychol., 1920, 4, 97-104.

Prince, M. The Unconscious. New York: Macmillan, 191 5. Pp. 549. °

Rosanorr, A. J. A Theory of Personality Based Mainly on Psychiatric Experi-
ence. Psycuox. BuLL., 1920, 17, 281-299.

Rucu, G. M. A Preliminary Study of the Correlations between Estimates of
Volitional Traits and the Results from the Downey “Will Profile.” J. of Appl.
Psychol., 1921, 5, 159-162.

Ruce, H. Is the Rating of Human Character Practicable? J. of Educ. Psychol.,
1921, 12, 425-438.

Soutuarp, E. E. On Descriptive Analysis of Manifest Delusions from the Sub-
ject’s Point of View. J. of Abnor. Psychol., 1916, 11, 189-202.

SoutHarp, E. E. On the Application of Grammatical Categories to the Analysis
of Delusions. Phil. Rev., 1916, 25, 424-455.

Spautpinc, E.R. The Réle of Personality Development in the Reconstruction
of the Individual. J. of Abnor. Psychol. & Soc. Psychol., 1921, 16, 97-114.
Tanner, A. Adler’s Theory of Minderwertigheit. Ped. Sem., 191 5, 22, 204-217.
Terman, L.M. The Intelligence of School Children. Boston: Houghton-Mifflin,

1919. Pp. 313.

Terman, L. M. The Army Rating Scale. Psycuot. Butt., 1918, 15, 177-187.

Tuornpixe, E. L. A Constant Error in Psychological Rating. J. of Appl.
Psychol. 1920, 4, 25-29.

. Rating Charts for Students and Teachers. Chicago: Univ. Chicago Bookstore, 1920.
61.

Wasusurn, M. & Oruers. Directed Egocentric Reactions. Amer. J. of
Psychol., 1919, 30, 300-302.

Watson, J. Psychology from the Standpoint of a Behaviorist. Phila.: Lippincott,
1919. Pp. 420.
63.
64.

65.
. Wexs, F. L. The Systematic Observation of the Personality in its Relation to

67.
68.
69.

70.
71.

PERSONALITY AND CHARACTER 455

Weaver, E. W. Occupational Fitness and Promise. Brooklyn: Vocational
Guid. Assoc., 1919.

Wess, EF. Character and Intelligence. Brit. J. of Psychol., Monog. No. 3, 1915,
1-99.

Weis, F.L. Mental Adjustments. New York: Appleton, 1917. Pp. 320.

the Hygiene of Mind. Psychol. Rev., 1914, 21, 295-333.

Weuts, F. L. Association Types and Personality. Psychol. Rev., 1919, 26+
371-381.

Wetts, F. L. Association Experiments Bibliography. Psycuot. But., 1916,
13, 146-147.

Wurtz, W. A. Mechanisms of Character Formation. New York: Macmillan.
1916. Pp. 342.

Woopwortn, R.S. Psychology. New York: Holt, 1921. Pp. 570.

Yerkes & La Rus. A Study of the Self. Cambridge: Harvard Univ. Press,
1914. Pp. 24.
MENTAL WORK! \
BY E. S. ROBINSON
University of Chicago

A sharp line of distinction between mental and physical work
is, as one might expect, still undefined. Thorndike (81, pp. 1-7)
not without some hesitancy and reservation has suggested that
mental work is that performed by the connection system. Watson
has subjected this statement to criticism (92, p. 350). According
to his own suggestion, what we commonly think of as mental work
differs from physical work in regard to the muscle groups employed.
Mental work is for him mainly laryngeal work.

Thorndike (81) and Dodge (24), although approaching the
problem from essentially different angles and although differing
very much in their modes of experimentation, have a good deal in
common in their contributions to the theory of mental fatigue.
Thorndike? has divided theories of mental fatigue into two classes
(pp. 119 ff), the mechanical and the biological. According to the
mechanical theories, mental fatigue is brought about by the con-
sumption of a stock of potential energy. Such theories he objects
to largely on the grounds that efficiency losses as shown by the
mental work curve are, because of their extreme irregularity, more
easily interpreted in terms of such factors as changing interest,
distraction by headache, and the like than in terms of energy
consumption in its simple sense. According to the biological type
of theory, which Thorndike favors, work without rest decreases in
efficiency largely because “it becomes less satisfying (1) by losing
the zest of novelty, (2) by producing ennui, a certain intellectual

1 This review will discuss developments in the field of mental work from 1912 to
the present time. Papers concerned mainly with physiological aspects of work have
not been considered. Undoubtedly the psychologist will find much in them that is
interesting and valuable, but a review such as this must stop somewhere. Studies of
work in industry are usually concerned with operations, which, if we make a common
sense distinction, are muscular rather than mental; for that reason they have not
been considered in any detail. On the other hand, where psychological investigations
have included data on obviously sensori-motor processes, these data have been con-
sidered. The material reviewed here, while not selected with any too rigorous
consistency, does represent a fairly unified line of study.

2 Thorndike’s conclusions here are based upon his researches dating back to 1900.

456
MENTAL WORK 457

nausea, sensory pains and even headache, and (3) by imposing
certain deprivations—for instance, from physical exercise, social
intercourse, or sleep” (p. 122). Dodge has met the energy-con-
sumption theory of mental fatigue in a manner not unlike that of
Thorndike. Although in an earlier paper (23) he recommended an
attempt to measure amount of mental work in terms of a metabolic
unit, his interest has apparently changed somewhat. He now
insists upon the importance of relative fatigue which “is not a mere
limitation of human energy. It is not exhaustion, but prevents
it” (p. 112). He thinks it unlikely that ordinary decrements in
the mental work curve are results of true fatigue processes in
nervous tissue. ‘They are more apt to be due to incidental inhibi-
tions. The views of both Thorndike and Dodge have put a whole-
some emphasis upon the complexity of the causes underlying
fluctuations in the curve of work. They have shown quite dis-
tinctly what mental fatigue is not and they have suggested the
importance of studying not only the decrements in efficiency of
given functions, but also the many forces, incentives, or what you
will, accelerating those decrements.

Whereas Thorndike and Dodge have tried to give analyses of the
phenomena commonly known as mental fatigue with a view toward
correcting the false analogies drawn between it and fatigue in the
narrower, physiological sense, Watson (p. 349) has recommended
that the concept of fatigue be banished from psychology. Evi-
dently he does not mean that loss of efficiency caused by continuous
work should not be studied, for he treats this subject at some length.
Evidently, too, he does not mean that such a question as that of
incentives should be left out of account, for he suggests that every
ascertainable aspect of the situation in which work is done may well
be noted. His main quarrel, it seems, is with physiological, or per-
haps he would say synaptic, theories of efficiency. “The point we
would make in doing behavior work,” says Watson, “is to leave out
of discussion all those factors which are not touched or approached
by the problem in hand; for example, what possible good does it do
when discussing bricklaying or subvocal arithmetic to guess at what
goes on in the synapse, in the efferent or afferent leg of the reflex
arc or in the muscle itself?” (p. 352). This line of attack upon the
- customary treatment of mental fatigue differs from those of Thorn-
dike and Dodge. They criticize the physiological theories of mental
fatigue because they feel that these theories are erroneous or, at
least, incomplete. Watson, apparently, criticizes them not so
458 E. S. ROBINSON

much because they are erroneous or incomplete as because they are
physiological and because they are theories.

Muscio (66) quite recently has again pointed out the fallacy
of calling all decrements in efficiency growing out of work effects of
fatigue. Like Watson, he would banish the term from scientific
discussion.

Methods of measuring fluctuations in mental efficiency have
not undergone any very radical changes during the past decade.
Investigators of the nature of the work curve have naturally em-
ployed the continuous work method. Most of their work has been
done with arithmetical activities. Investigators of the effects of
drugs, hour of day, loss of sleep, fasting and the like have usually
employed the testing method. In general there has been a decline
in the use of such tests of efficiency as esthesiometry, sensory acuity
and spontaneous tapping rates.!

While no radically new methods have been introduced into this
field, there has been a constant refining of technique. Greater care
has been taken to secure adequate controls over the attitudes of the
reagents studied. Rivers, somewhat before the period covered by
this review, made much of the faulty methods of most of the earlier
investigations of fatigue, the effect of drugs, etc., and experimenters
during the past ten years have taken many new precautions against
the vitiating influences of psychological attitude.

Proposals to measure amount of mental work done in terms of
some physiological change are not new. There have, however,
been some rather interesting suggestions along this line. Dodge
(23), for example, has pointed out the close relationship between
pulse rate and general metabolic conditions and has suggested
pulse rate as a measure of mental work done. His paper on this
subject contains some data on the relationship between pulse rate
and the performance of a variety of intellectual tasks. Urban (go)
criticizes Dodge’s program on the grounds that evenif pulse rate were
an exact measure of metabolism, only the bodily accompaniments
of mental work would be measured by it. Such a criticism could
not be answered except on metaphysical grounds. Lehman (54)
also desires some measure of mental work in physiological terms. He
believes that mental fatigue is accompanied by an increased amount
of exhaled CO.”

For a discussion of these tests as measures of mental fatigue, see: OFFNER?
Mental Fatigue translated by G. M. Whipple.

_ *The reader may recall the work of Benedict, Atwater, and Capenter on this
point.
MENTAL WORK 459

Another important paper from a methodological standpoint is
a recent one by Thorndike (84) on the curve of satisfyingness.
Here emphasis, which has during the past ten years been everywhere
else, turns back again upon “feelings of fatigue,” or at least upon
data obtained from purely subjective observations. The subjects
of this experiment periodically rated the “satisfyingness” of a
continuous task.

Tue Curve or Work

Perhaps the most important paper on the curve of work from a
critical and theoretical standpoint is one of Thorndike’s (85). He
attacks the analyses which Kraepelin and others have made of the
course of efficiency during continuous work. As a result of their
analyses a number of features or elements stand out as characteristic
fluctuations in efficiency. These are practice effect, fatigue effect,
“warming up” effect (Anregung), adaptation (Gewohnung), initial
spurt, end spurt, spurts after fatigue (Ermidungsantriebe), spurts
after disturbance (Storungsantriebe) and the rhythm of attention.
Thorndike is inclined to believe that these features of the curve of
work have been much overemphasized. ‘‘The most important fact
about a curve of efficiency of a function under two hours or less
continuous maximal exercise is that it is, when freed from daily
eccentricities, so near a straight line and so near a horizontal line”
(p. 188). It seems to be his main contention, however, not that
warming up, initial spurt, and the like are fictitious, but that they
are conceptions which should be empirically investigated, rather
than conceptions which can profitably be used as principles ex-
planatory of work curve irregularities.

Chapman (16) found clear evidence for an initial spurt in
continuous adding. Ina later paper by Chapman and Nolan (19)
it appears that initial spurt is more marked where the work is
divided into half-minute periods than where it is divided into two-
minute periods. This suggests that initial spurt may be so short
in length that it can easily be concealed where average scores for
relatively long successive periods are considered. Phillips (71)
obtained evidence for an initial spurt in a ten-minute period of
arithmetical work. During the first minute his subjects performed
from 6 to 12 per cent. more efficiently than they did, on the average,
during the remaining 9 minutes. Thorndike (83) has also published
some positive evidence in regard to initial spurt (his subjects wrote
from memory or with reference to a key answers to multiplication
problems). For a minute or so at the beginning of the work the
460 E. S. ROBINSON

subjects maintained a speed which they soon abandoned. Thorn-
dike does not interpret this as evidence of a general fact of initial
spurt. He thinks that, since the earliest work is not immediately
preceded by any work like it, it is, therefore, less open to confusion
effects than the later work.

“Warming-up” also seems to occur under proper conditions,
i.e., where each day’s work is in a sense a learning process. Starch
and Ash (79), for instance, have results on two and one half hours
of adding which indicate a rise in efficiency as measured by both
speed and accuracy during the first 25 to 35 minutes. End spurt,
too, it is generally admitted, may occur when conditions are proper,
the sine qua non being knowledge on the part of the subject as to
when the work is to cease.

A number of studies have contributed to our knowledge of the
amount and nature of the losses in efficiency due to the continuous
operation of various specific functions. Arai (3) performed mental
multiplications (four-place by four-place numbers) continuously for
as long as twelve hours. At the end of such a period it took her
100 per cent. longer to do one problem than it had at the beginning
of the period. Groups of subjects engaging in such work for two-
hour periods took 24 per cent. longer per problem at the end than
at the beginning of the period.

Garth (32) found that children, adding for 28 minutes or 42
minutes, suffer an efficiency loss of fromotos5 percent. Thorndike
(82) discovered that the loss from four hours of poetry writing is a
matter of quality and satisfyingness rather than of quantity of
work done. Painter (67) shows that one’s ability to work may give
out suddenly. After a hard day’s work he multiplied mentally
four-place by four-place numbers from 11 p.m. until 3 a.m. At
the end of the period he was unable to continue this work. He
was also unable to perform easier multiplications. Vernon (91)
claims that among women workers in English munition plants the
12-hour day caused 2.5 times as many accidents as the 10-hour
day. If this conclusion is sound it is another illustration of a
sudden acceleration of the fatigue process. Poffenberger and Tall-
man (73) suggest that mental fatigue may proceed at a rate obscured
by the conditions of most experiments. They found that the first
half of a test, the total length of which is less than one minute, is
done more quickly than the second half. But the repair process
may also be so rapid that unnoticed periods of relaxation within

the ordinary work period may hide an actual alternation of fatiguing
and recovering.
MENTAL WORK 461

The extent to which the continuous operation of a function
affects its efficiency is a matter not only of the function involved
but also of the method of testing it. Morgan (63) had subjects
spend four hours at studying the English equivalents of 850 German
words. Much more of the material studied early in the sitting was
recognized at a subsequent test than of that studied late. Learning
scores and relearning scores on the other hand show no such clear
distinction between the efficiency of the earlier and later parts of
the study period.

PuysioLoGicaL Errects or Menta, Work

The effects of so-called mental work upon so-called physiological
activities are still far from being worked out to any complete extent.
In 1913, Dodge, as already indicated, reported certain correlations
between variations in pulse rate and the performance of certain
kinds of tasks. Arai’s results (3) show a decrease of pulse rate
with long-continued mental multiplication. Patrizi (69) found
some correspondence between volumetric changes in the arm and
auditory stimulation. These changes seem to be controllable by
some subjects. According to another experiment performed by
Patrizi (68) it appears that the ergographic work of some subjects
is benefited and of others interfered with by an accompanying
mental task. Zondek (96) has published a paper in which he shows
that a brief period of writing fatigues the muscles of the legs.
There is some doubt as to the degree to which writing should be
considered a mental task. The results which I have mentioned
are typical of those in this particular field. As far as they go they
are definite enough, but their interpretation and explanation is
well nigh impossible largely because the mental activities with
which we are interested in dealing are not at the present time
susceptible to anything like a complete analysis from a physiological
standpoint.

SuBJEcTIVE EFFECTS

_ Reports of feelings of fatigue have for the most part come as
by-products of studies concerned mainly with more objective data.
Arai (3), for instance, speaks of a positive but slight correlation
between feelings of fatigue and production records. G. E. Phillips,
(quoted in 65) obtained a similar result. JIoteyko (41) makes the
point that sensations of fatigue, like pain sensations, do not follow
Weber’s law. In these realms sensation values increase more
rapidly than stimulus values. Robbins, Smith and Washburn (75),
462 E. S. ROBINSON

following up an earlier Vassar experiment, show that a prolonged
series of judgments on the affective values of colors tends to lower
the affective values of colors for the subject judging. Berliner (10)
tested the ability of subjects to hold memory images. The experi-
ments were conducted in the morning before the day’s work and
in the evening after the day’s work. Her subjects held the images
considerably longer at the morning sittings. Thorndike’s paper
(84) on the curve of satisfyingness has already been mentioned.
In one experiment five subjects worked at the same task for one
sitting of four hours. At the end of successive twenty-minute
periods the subjects rated the satisfyingness, tolerability or zest of
their task on a scale ranging fromo toro. According to the results
one becomes bored with a continuous task at a much greater rate
than one loses in speed or accuracy of performance. Muscio (65)
reports the estimates of their own “tiredness” supplied at different
hours of the day by a group of 20 women medical students and by a
group of 15 women typists. In general the diurnal curves fall as
the day advances, but for both groups there is a sharp rise in feeling
tone just after luncheon.

Errects oF OnE Form or Mentat Work upon ANOTHER

Since rest is at best a term of only relative significance, the
problem of the effect of rest upon work is, when more exactly
formulated, a problem of the effect of one form of mental activity
upon another. Dockeray (20) studied the effects of physical work
upon mental work with very irregular results. Arai (3) states
that the degree to which loss of efficiency due to the continuous
exercise of one function operates deleteriously upon other functions
is dependent upon the extent of loss in the first function.

More definite findings appear in the studies of Ash (6), Chapman
(17), and Fernberger (29). Ash found that adding has a marked
influence on one’s ability to control reversible perspective. He
avoids an interpretation of this fact in terms of consumption from
a common energy supply by assuming that it is not the amount
but the control of energy which is affected. In Chapman’s experi-
ment it was found that half-minute periods of adding were not as
efficient when continuous for three minutes as when half-minute
periods were alternated with cancellation or rest. He explains this
fact in terms of interference. According to Fernberger a half-hour
period of translating dificult German has no appreciable influence
upon the judgment of lifted weights. Where dynamometric or ergo-
MENTAL WORK 463

graphic work was done with the muscles used in lifting, however,
there was a considerable effect. The interval of uncertainty, for
example, increases by 30 per cent. and there is a marked tendency
for the second weight to be overestimated.

Rest Pauses

Considerable work has been done on the optimal distribution
of exercise for habits still in the process of formation, but little
that the reviewer is aware of has been done on the optimal distri-
bution of effort for fully acquired habits. This is true at least for
habits which would ordinarily be called mental. The effects of
various distributions of ergographic work have been studied and
some attempts have been made to establish optimal work and rest
periods for industrial workers. Jones (46) describes two investi-
gations of this sort. “Mr. C. E. Knoeppel started with workmen
who, at their own gait, had been producing 16 pieces per hour.
By establishing a 25-minute working period and a 5-minute rest
period, he obtained 18 pieces. By changing to 17 minutes of work
and 3 minutes of rest, the output rose to 22 pieces perhour. Finally
by arranging a 10-minute work period and a 2-minute rest period,
production became 25 pieces. In another case a record of driving
1,600 rivets per day was obtained, the previous performance having
been 600 per day, Mr. H. F. Stimpson establishing rest periods of
2 minutes between each Io rivets, thus devoting 320 minutes, or 5
hours and 20 minutes out of the 10-hour day to rest, and employing
a schedule of 1 3/4 minutes of work and 2 minutes of rest” (p. 222).
Similar investigations would probably yield as interesting results
for the so-called mental types of work.

In 1913, Dodge (22) reported some experiments on the wink
reflex which might well be mentioned here. In disagreement with
the earlier claims of Zwaardemaker and Laws his results show no
absolute refractory phase for this reflex. As two stimuli (sounds)
are presented closer and closer together in time, the percentage of
double reactions gradually becomes less.

DisTRACTION
Distraction has been studied mainly by the reaction-time
method. Asa result of the work of Cassel and Dallenbach, Evans,
Morgan, and Todd, a number of interesting facts and problems
have been uncovered. Distraction may increase or decrease the
speed of reaction, depending upon the temporal position of the
464 E. S. ROBINSON

distraction (15, 88). For example, distractors given before the
reaction stimulus may increase reaction time, while the same
distractors given with the reaction stimulus may decrease reaction
time (88). Apparently the most effective distraction is sound.
The effectiveness of a distraction is also greater if it is of the same
sensory modality as the reaction stimulus (28). The power ofa
given distraction may be reduced by increased effort on the sub-
ject’s part (63), or by a process of habituation (15, 28). On the
whole, continuity of the distraction favors adaptation to it (15).
When once a given distraction has been adapted to, that adaptation
will hold over to some extent even to conditions where the stimulus
to reaction is changed (28). Certain of these findings are true
in an undoubtedly wider sense than others. All of them suggest
problems and hypotheses in regard to mental work in general.

INCENTIVES

The importance of incentives is pointed out again and again in
studies of mental work. Investigators of the effects of drugs,
ventilation and the like have endeavored to control these factors
as well as possible. The usual procedure has been that of supplying
the maximum of incentives practicable under the circumstances of
the experiment. This has been done by directing the subjects to
work as hard as possible, by giving them access to their records
and in a number of similar ways (see, e.g., 80). Writers on indus-
trial work of various types also make plain the importance of incen-
tives, but although they occasionally offer empirical data, it is
usually too complicated to invite any confident interpretation.

Straightforward studies of the incentive problem, as such, are
rare. Arps (5) found that subjects do ergographic work somewhat
more efficiently when aware of their records. Chapman and Feder
(18) found that school children do better in certain tests when
offered prizes and when made aware of their results, and Allport
(2) found that even on simple tasks subjects are influenced by their
social surroundings.

It is natural to wonder why such obviously important factors
in efficiency as incentives have not been subjected to thorough
experimental investigation. The answer lies in the extreme com-
plexity of the incentive problem. There are probably as many
incentives affecting the quality and quantity of mental work as
there are kinds of likes and dislikes in human life. And every
student of the subject recognizes that these are not only hard to
MENTAL WORK 465

isolate, but also hard to define. Rivalry, a very evident incentive,
is one which can be experimented with in a way, but an investigator
working with such a factor would require a great deal of insight,
or at least of hardihood, to interpret his results.

Loss or SLEEP

Smith (78) carried out a rather elaborate investigation on the
effects of loss of sleep. Five tests were employed: (1) the Mc-
Dougall dotting machine, (2) an associated words test, (3) reversible
perspective, (4) tapping, and (5) a nonsense syllable test. Follow-
ing a period of insomnia the subject showed an increased capacity
for tasks requiring concentrated attention, 7.¢., the dotting machine
and associated words tests. Later a period of lowered efficiency set
in and records remained below normal for some days. A further
important finding is brought to light by the results of an experiment
where a period of insomnia was interpolated before efficiency,
reduced by a previous period of insomnia, had returned to normal.
Fatiguing the subject who was already in a fatigued state tempor-
arily increased her efficiency. Subsequently, however, there set
in a prolonged period of subnormality. Miss Smith suggests the
possibility that “the final result of the fatigue toxins is to produce
a substance allied to the drugs with a stimulating action . es
(p. 344). Some of-the fatigue products might thus act to make the
operation of the nervous system temporarily more efficient.

While these results and tentative interpretations are interesting
and important two facts, at least, it seems render them unaccept-
able at their face value. In the first place only one subject was.
employed. Although the effects of fatigue are fairly consistent
throughout her records, it is not at all improbable that these effects
would differ for different subjects. Ifa group of subjects had been
used, Smith’s most striking finding, i.¢., the stimulating effect of in-
somnia, might have appeared only in certain individuals. At any
rate, as she herself suggests, the experiments need repeating on a
larger scale. In the second place, Smith, in interpreting her results,
has emphasized physiological explanation to an extent which is
hardly justified. It does not seem at all unlikely that the positive
effects of insomnia were due to the subject’s greater relative interest
in the general experimental situation on the day or days immedi-
ately following the insomnia.

Patrick and Gilbert in their paper of 1896 (Psychological Review)
noticed that even a moderate amount of sleep is capable of com-
466 E. S. ROBINSON

pensating for a condition of unusual fatigue, but Smith’s experi-
ments indicate that “the time taken to return to a normal condition
after the loss of but a few hours’ sleep is disproportionately great;
and this return is gradual but irregular” (p. 349).

Smith’s subjective reports are in essential agreement with those
of Patrick and Gilbert. There were disturbances varying from
slightly increased irritability to actual hallucinations.

FastTInc

The effects upon mental capacity of going without food are by
no means clearly made out. This may be due to a number of
causes. The effects may be very complex, the tests so far employed
may be lacking in delicacy, and important after-effects have un-
doubtedly escaped notice due to the fact that adequate “after”
tests on this condition have yet to be made.

Langfeld (50) made daily examinations of Agostino Levanzi, a
lawyer of Malta, who underwent a 31-day fast at the Boston Nutri-
tion Laboratory of the Carnegie Institution of Washington. Due
to unavoidable circumstances there were only three days of testing
before the beginning of the fast and no adequate tests at the com-
pletion of the fast. A considerable number of tests were used and
from their results Langfeld concludes that: “In a word there was a
loss in muscular strength due probably to loss of tissue, a possible
gain in sensory acuity and a decided increase in the efficiency of all
the central processes” (p. 48). (The efficiency of the central proc-
esses, of course, is supposedly measured by the more complex
intellectual tests.) The loss in muscular strength is obviously a
fact caused by fasting. The improvement in sensory acuity and
‘central processes” is not so clearly so. Possibly in certain of the
sensory acuity tests and surely in the “‘central processes’ tests
there is a rise due to practice which is to be expected, fast or no fast.
The only conclusion to be drawn, then, in respect to these tests is
that practice effects in them were not obliterated by the period of
fasting. Such effects may have been either positively or negatively
accelerated, or they may have been unmodified. The subject’s
belief in the advantages of fasting was such that introspective
reports could not be relied upon. There were some indications of
irritability which may have been caused either by the fast or by the
monotony of the subject’s life under the conditions of the experi-
ment.

In a paper entitled “Individual and Sex Differences Brought
out by Fasting” (57) Marsh reports an experiment in which he
MENTAL WORK 467

and his wife were subjects. During the first week feeding was
gradually reduced from normal, during the second week no food
was taken, and during the third week feeding was gradually in-
creased to normal. A number of physical and simple and complex
intelligence tests were given daily during the experiment. No
control data (from “before” and “after” tests) are presented.
The results indicate a positive correlation between food intake and
vitality and strength. There is no clear-cut finding in regard to
the intelligence tests. What so-called sex differences appear are
naturally of no significance, because there is not the slightest
guarantee that similar or greater differences would not under the
same circumstances appear in the records of two subjects of the
same sex. Subjective reports indicate pronounced emotional dis-
turbances due to the fast. These varied from a condition of apathy
to one of extreme excitability.

In examining such studies as those on fasting and loss of sleep
one can not help but be struck by the fact that, whereas effects
revealed in test scores are fairly meager, those revealed in intros-
pective reports are often pronounced and definite, even if unsys-
tematic. Indeed, it is not unlikely that further studies along these
lines will be greatly aided if they contain in their programs some
provisions for controlled (to the extent to which they can be con-
trolled) introspective observations. Of course another line of
procedure lies in the direction of more refined tests.

Errects oF Drucs

Rivers, as has been noted, emphasized the lack of adequate
controls upon the subject’s attitude in the earlier studies of work.
Watson (p. 360) points out three additional limitations present in
many of the experiments on the effects of drugs. (1) “only in rare
cases has a function been practiced sufficiently in laboratory studies
before control tests were introduced to gauge the effects of drugs,
continuous exercise, loss of sleep and other factors.” (2) “Too
narrow functions have been studied—rarely has any of the work
been done upon functions used in daily life... .’’ (3) “the
functions have been followed through for entirely too short periods
for wide generalizations to be made.”’ In general, research workers
have become increasingly careful to avoid these sources of error,
but it is frequently impossible to avoid all of them. Take for
example, the case of the subject’s attitude. In some cases it is
practicable so to disguise a given drug that the subject will not
468 E. S. ROBINSON

know whether he has taken caffeine, say, or sugar of milk. Experi-
ments with alcohol and tobacco, however, present more difficult
problems. The flavor of alcohol is not easily disguised and, if
doses of any size are employed, its almost immediate subjective
consequences are quite apparent. And a really adequate experi-
ment on tobacco smoking should provide for a control in the form
of smoke from which all effective elements have been removed. It
is likely, however, that where completely disguised doses of a given
drug are impossible, longer series of measurements with more
subjects eliminate to a considerable extent differences due merely
to attitude.

Alcohol.—Experiments reported in 1911 or earlier (48) indicate
that alcohol usually acts detrimentally on mental functions. This
is quite certain where complex processes are involved and where
the doses are moderate or large. A possible exception appears in
a study by Langfeld (51). He found that alcohol reduced certain
association and reproduction times without affecting the accuracy
of these processes in any certain manner. While the results from
very small doses or in very simple processes are doubtful there is
at least no clear evidence here for either permanently or temporarily
increased efficiency. It seems quite probable that where small
doses failed to cause bad effects it was due either to the extreme
variability of the function studied or to the fact that the subject
put forth an increased effort to offset a lowering capacity of which
he was aware. From these earlier experiments it also appears that
the effects of alcohol are more evident in the quality than in the
quantity of work.

Later work has on the whole substantiated the conclusion that
alcohol has a detrimental effect upon mental operations. Frank-
fiirter (30) working with himself as subject, found that 20 c.c.
doses of alcohol decreased the accuracy of his typewriting and also,
be it noted, his speed. While Lange and Sprecht (49) found that
the visual and auditory stimulus thresholds were lowered by small
amounts of alcohol, the difference thresholds for the same senses
were raised.

Dodge and Benedict (25, 26) conclude from the results of an
extremely careful series of experiments that alcohol is a depressant
rather than a stimulant. This is shown in the following tabulation
of their results quoted by Hollingworth and Poffenberger (40, p.
172):
MENTAL WORK 469

Per Cent.
Increase of latent time of knee-jerk...............00.00 0000s 10
Decrease in the thickening of the quadriceps muscle.......... 46
Protective eyelid reflex, latent time increased................ 7
Extent of eyelid movement decreased... ........0..000 ee eee 19
Eye reactions, latent time increased......... sitter eee ae 5
Speed of eye movements decreased. ...........0cee cece eeeee II
Sensitivity to electric stimulation decreased...............005 14
Speed of finger movements (tapping) decreased.............. 9

Miles (61, 62) has in two papers published findings which in
general verify those of Dodge and Benedict. In spite of such
results as those of Lange and Sprecht in regard to visual and
auditory stimulus thresholds, we are safe in concluding from the
thorough and rigorous experiments of Dodge and Benedict, and
Miles, that even in the realm of comparatively simple processes
alcohol is mainly a depressant and that the detrimental effects of
alcohol (this conclusion finding support from Frankfirter) are clear
in respect to speed as well as in respect to quality of work.

McDougall and Smith (60) obtained results demonstrating that
alcohol usually has a deleterious effect upon mental efficiency.
Varying doses of alcohol increased the number of errors in the
dotting machine experiment as follows:

Dose Increase in Errors, Per Cent.
NOC. Cesena: aise oe Gilsaite t's; oS wie alls gia Aenea reas a eae ne eensoateeearenmie wae 21
TG Gear igus see dS oes sad ede. s sata d eine perme oe 42
DO, CC ds cites aid enemies en C aaatanes Mon eG lain davetbelet teh an ss 39
DG CiCriccdiecclsltdie viarstcaianest dG POS FSR MEMRMIEEES 44 5s ROR OaS 113

A memory test also showed that alcohol lowers efficiency.

There is an exception to the generally deleterious effect of
alcohol in the records of McDougall and Smith which is very
interesting. It appears that when alcohol was taken at any time
during a period of several days following a loss of sleep and while
efficiency was dropping as a result of that loss of sleep, then alcohol
still farther reduced efficiency. When, on the other hand, alcohol
was taken during the period in which the subject’s efficiency,
lowered by loss of sleep, was gradually rising, then alcohol in-
creased efficiency. These generalizations, if they can be substanti-
ated, are obviously of first rate importance. That they need
substantiation is evident from the fact that they seem to be based
upon the performance of one subject.

The same investigators studied the effects of two different
concentrations of alcohol. Twenty c.c. of absolute alcohol diluted

‘
a
470 E. S. ROBINSON

to 60 c.c. water increased the errors in the dotting machine test
160 per cent. 20 c.c. of absolute alcohol diluted to 200 c.c. of
water, on the other hand, increased the errors by only 22 per cent.
They also report that the effect of alcohol is much less evident if
the alcohol is taken with a meal. When 30 c.c. was taken under
such circumstances it had little or no influence upon efficiency.

The efficiency reducing capacity of alcohol is farther verified
by Jorger (47) who found that this drug increases free association
time and lowers the quality of the response and by Arlitt (4) who
tound that alcohol affects the maze-learning ability of white rats
to which it is fed and also that of their offspring.

The old notion, still popular in some non-scientific quarters,
that alcohol may have at least temporary beneficial effects probably
grew out of a misinterpretation of a very real fact about alcohol,
namely, that the subject who has taken a moderate amount of it
is inclined for a time to over-rate the excellence of his own perform-
ance. McDougall and Smith report this tendency and popularly
we are aware that many a story of keen wit becomes dull when its
audience is free from the effects of alcohol.

The possibility remains that there are important individual
differences in the degree of susceptibility to effects of alcohol. In
fact the reality of such differences hardly needs confirmation. But
whether some individuals are actually stimulated by a drug which
depresses others is still another matter. Dunlap (27) tested certain
eye-hand codrdinations of two subjects each of whom were given
within about one hour 7 oz. of rye whiskey. One of these subjects
became obviously drunk and his codrdinations clearly disturbed.
The codrdinations of the other subject seemed to improve a little
during the experiment, and he felt no ill effects until a vomiting
spell 8 hours after taking the whiskey. The data are too meagre,
however, to justify the conclusion that the second subject was
made even temporarily more efficient by the whiskey.

Caffeine.—Most of the evidence which we have upon this subject
indicates that moderate amounts of caffeine (amounts such as are
found in the ordinary cup of tea or coffee) increase the efficiency of
mental activity, and there is little reason to believe that this rise
in efficiency is followed by a measurable decrement. The effects of
larger doses are sometimes unfavorable, depending somewhat upon
the operations considered and the subjects.

Hollingworth (38) has presented the more important experi-
mental facts reported prior to 1912. The following table gives the
main results of his own investigations;
MENTAL WORK 471

 

 

 

 

St. = stimulation © = no effect Ret. = retardation
Test Small | Medium | Large |Secondary
Doses Doses Doses | Reaction
¥,. Tapping: scuoyeee ovess Se ddaddediees St. St. St. None
2. Three-hole: scccsicsceesseeosea ee ed St. ° Ret. None
3. Typewriting
(a) SPCed is. des scent aaeaaeies ays St. ° Ret. None
(GO) EXrOTs 536-204 ose bawaediada ns Fewer for all doses None
4. Color naming. ...............0005 St. St. St. None
Gx OPPOSitesiwise: dsiueisdga dahl’. hess cied St. St. St. None
6,. Calculation osc 06s savnsens exeacass St. St. St. None
7. Discrimination reaction time........ Ret. ° St. None
8. Cancellation.............-..2. 000+ Ret. ? St. None
g. S-W illusion... ...........0.e eee ° ° °
10. Steadiness............-..2- cere eee > Unsteadiness None

 

 

 

 

The sleep of the subjects was generally untroubled except where
large doses were taken. In that event such symptoms as headache,
dizziness, feverishness and irritability were also liable to be present.
The degree to which sleep and feelings of well-being were affected
by caffeine varied with the body-weight of the subjects. The
heavier ones were less affected.

Briefer experiments by Langfeld (51) and Tolman (89) support
Hollingworth’s findings. According to the former caffeine decreases
association time, and according to the latter small doses of caffeine
(1.5 grains) increase one’s ability to memorize and, at the same
time, lower one’s susceptibility to retroactive inhibition. In oppo-
sition to the general trend of these results we have those given by
Schilling (77). He found that simple reactions to an auditory
stimulus are, if anything, retarded by 5 grain doses of caffeine. It
should be noted, however, that such doses are relatively large.
Lashley (52) reports that caffeine lowers the learning capacity of
white rats.

Hollingworth points out (p. 166) that quantitative results from
investigations of the effects of caffeine in its pure form cannot be
compared uncritically with its effects when administered in tea,
coffee and other beverages. Hollingworth’s experiments show that
caffeine has a greater effect when taken on an empty stomach.
This statement may well be kept in mind in regard to other drugs.
The effects of alcohol (see McDougall and Smith) for instance,
vary considerably according to the form of administration.

Tobacco.—It has been established for some time that excessive
smoking is injurious to the physical organism. Just what amount of
472 E. S. ROBINSON

smoking should be thought of as excessive and just how injurious a
given excess is depends within rather wide limits upon the individual
smoker. Certain correlations have also been pointed out between
tobacco smoking and such measures of intelligence as school marks
(12). With practically no exceptions groups of non-smoking stu-
dents make better marks than groups of smokers. But the earlier
disposition to interpret this fact as a sign that tobacco smoking
has a measurably detrimental effect upon intelligence has been
checked by more careful analysis. It is quite possible that where
small boys take up smoking they do so because they lack certain
social inhibitions, and the lack of these same or closely related
inhibitions may also be a cause of their disregarding scholastic
duties. Among older boys perhaps the socially more intelligent
are in a position where the tobacco habit is more easily cultivated.
And they may also be in.a position where there are a maximum
number of distractions from purely scholastic activities. ‘Thus one
aspect of intelligence might be the cause of both tobacco smoking
and low scholarship. In other words the true state of affairs is
far more complex than that suggested by the statement of a single
causal relationship between tobacco smoking and intelligence or
even tobacco smoking and scholarship.

Experimental studies of the effects of tobacco smoking upon
mental functions indicate in general that, whatever the amount of
this effect, itis detrimental. While Berry (11) found that he added
slightly more accurately or more rapidly after smoking a cigar,
Bush (14), Johnson (45) and Froeberg (31) obtained some evidence
of an advantage in the other direction. As far as these experi-
mental studies go they indicate that the effects of tobacco smoking
on mental efficiency are very slight. Before taking such a conclu-
sion seriously, however, one should examine some of the limitations
of its factual support:

1. During these experiments the smokers used much less tobacco
per day than does the ordinary user of the drug.

2. The effects considered are those accruing from small and
isolated amounts of tobacco smoking rather than from the relatively
continuous indulgence characteristic of smoking in everyday life.

3. Although Froeberg had some slight success along this line,
no one has yet devised an adequate control dose (i.¢., a dose minus
any effective agents) of tobacco smoke. Consequently differences
of attitude are bound to arise between the “smoke” and “non-
MENTAL WORK 473

smoke” tests or groups for which it is impossible to make accurate
allowance. Such attitude differences are especially important where
the data do not point very strongly to the advantage of one con-
dition, as in the case of these experiments in tobacco smoking.

So far it has not been proved that tobacco smoking aids mental
efficiency. But for moderate smoking neither has the opposite been
clearly established. Experimental interest in the subject has prob-
ably been too general in nature. A comparison of the effects of
tobacco with those of no tobacco is an indefinite procedure. Better
a measurement of the effects of doses of tobacco varying in size or
in frequency. Studies might well be made of the development of
tobacco tolerance (a fact which most smokers know to be quite
real) as measured by mental efficiency, and also of the loss of that
tolerance.

Writers who wish to defend tobacco smoking or to seem non-
partisan to tobacco smokers sometimes argue that, whereas tobacco
smoking may have some detrimental mental effects, so has the
elimination of tobacco smoking a detrimental influence upon the
smoker who is not accustomed to doing without it. This of course,
holds good for any drug, and it is probably less true for tobacco
than it is for the more harmful drugs, such as alcohol, opium, etc.

Strychnine.—Poffenberger (72) has summarized the effects of
strychnine as measured up to 1914. Small doses of this drug are
said to improve the appetite and to give the subject feelings of
strength and optimism. Ergographic work is first positively and
then negatively accelerated by strychnine.

Lashley (52, 53) reports that proper injections of strychnine
increase the maze learning ability of rats as measured by the amount
of practice required to achieve complete learning. Retention is
apparently unaffected. In the experiments of McDougall and
Smith (60) one subject was given doses of from 1/15 to 3/15 grains
of strychnine at intervals. The effects were marked. The dotting
machine test showed a decrease in errors of 15 per cent. and memory
for lists of associated words was similarly improved. Strychnine
also increased the number of alternations in a reversible perspective
figure.

Poffenberger’s own findings are mainly negative. Ordinary
therapeutic doses (1/30 to 1/20 grain) interfered slightly with steadi-
ness, but tapping, aiming, and a wide range of intellectual activities
were unaffected. He explains his results as perhaps due to the
alleged fact that strychnine acts upon the cord and medulla to a
474 E. S. ROBINSON

greater extent than upon the higher centers. This suggestion does
not touch the lack of harmony between Poffenberger’s results and
those of other investigators, however, and Watson (p. 368) holds that
Poffenberger may not have employed suitable doses for the partic-
ular subjects tested. This is at least plausible when we note that
McDougall and Smith who did obtain positive results used 1/15
grain as their smallest dose, whereas Poffenberger’s largest dose
was 1/20 grain.

Other Drugs.—Results obtained by McDougall and Smith indi-
cate that the maximal effect from 40 to 90 minims of tincture of
opium appear between 75 and 115 minutes after taking. The first
effect as shown by the dotting machine test and by a memory test
is one of increased efficiency lasting from 30 to 40 minutes and
followed by a drop to a level considerably below normal. Alter-
nations of reversible perspective which are decreased by alcohol
and chloroform are increased by opium just as they are by strych-
nine. The amount of stimulation supplied by opium varies appar-
ently with the fatigue state of the individual. When a subject who
had suffered from loss of sleep was on a plane of efficiency below
her normal one the stimulating effect of opium was most marked.
When the subject was under the influence of opium the test task
seemed relatively easy somewhat as it had in the case of alcohol.
But where opium was the drug, objective efficiency corresponded
more nearly to subjective estimates.

Macht and Isaacs (55) report that doses varying from 1/16 to
1/4 grain of morphine act as stimulants, decreasing reaction time
and increasing accuracy of response. The initial period of stimu-
lation is shortest with the largest doses of the drug. Following
this is a period of depression during which efficiency is below normal.
Effects essentially unlike those of morphine are reported by Macht,
Isaacs and Greenberg (56) for such drugs as quinine, aspirin, phen-
acetin and salol. Reaction times here were lengthened; there was
no period of stimulation.

One may well conclude from all these researches that in the
future the action of few drugs will be considered understood until
their effects on mental activities have been investigated.

CoLor AND BriGHTNEss OF SURROUNDINGS
While there is, or has been, a good deal of speculation about
the effects of color and brightness upon human working capacity,
there is little in the existing literature (aside of course from that on
MENTAL WORK 475

the purely optical aspects of the situation) to warrant definite
generalization. Pressey (74) reports that his experiments show
some slight evidence for a speeding up of certain functions (tapping,
multiplication, and continuous reaction) with increasing brightness
of the subject’s surroundings. There is no evidence for a potency
of hue independent of brightness. Similarly Johnson (42) found
that subjects sort cards by touch better when working in a good
light than when working in darkness, even when visual aids to
sorting are lacking in both cases. The mere presence of light as a
stimulus may so increase general muscular tonicity that the organ-
ism’s capacity for work is increased. Burtt (13) made tests of
attention and reaction time on subjects in evenly and unevenly
illuminated sections of a city street and on subjects working under
analogous conditions in the laboratory. His results show greater
efficiency for the conditions of non-uniform illumination.

ATMOSPHERIC CONDITIONS

Psychological investigations have been made of the effects of
the COs content of the air, of temperature and of humidity. Bass
(7, 8), and Thorndike, Ruger and McCall (86) report that within
the limits of their experiments, ventilation by recirculation instead
of by introducing outside air has no noticeable deleterious effect
upon the intellectual processes. Thorndike and others (87) com-
pared the psychological effects of ventilation conditions varying
in regard to temperature, humidity, admission of outside air, and
circulation of the air. No differences in efficiency were noted
between work done under supposedly favorable conditions (68° F.,
50 per cent. rel. hum., etc.) and that done under supposedly unfa-
vorable conditions (86° F., 80 per cent. rel. hum., etc.) Stecher (80)
studied the effects of humidity, temperature being kept constant
at 75° F. Two humidity conditions were compared; (a) relative
humidity of 50 per cent. and (b) relative humidity of 20 per cent.
Both subjective reports and the scores from ten tests fail to show
any significant effects. ‘‘Psychology,’’ Miss Stecher concludes
(p. 85), “has given its answer, so far as it can in the present state
of the science, to the question as to whether the ventilation con-
ditions commonly found have any effect on the practical activities
of life. If this absence of demonstrable effect is due to a constant
adjustment of the organism that will eventually result in strain,
it is for physiology to trace any subtle, long-time ill-effects that
may have escaped the behavior tests.”
476 E. S. ROBINSON

The psychological effects of oxygen deprivation were very
carefully studied during the war (1, 9, 43, 44, 92) in connection
with the aviation service. Lack of oxygen was secured by a re-
breathing apparatus by means of which the subject rebreathes the
air from a supply tank, thus reducing the oxygen content, or by
means of a diluting apparatus. ‘These devices were standardized
with reference to a low-pressure tank in which air can be rarified
to correspond to any given altitude. In the British investigations
(21) the effects of oxygen deprivation were frequently determined by
tests after flights. Ability to stand oxygen deprivation varies con-
siderably. Of each 100 carefully selected American flyers only 61
are physically and mentally capable of attaining an altitude of over
20,000 feet with safety; 25 out of each 100 are physically and men-
ally unsafe at altitudes above 15,000 feet; and 14 out of each 100
are physically and mentally unsafe at altitudes above 8,000 feet.
Of psychological processes the more highly organized are impaired
first. Oxygen hunger has accumulative as well as immediate effects.
One hundred hours of flying at high altitudes may render a man at
first capable of effective work at 20,000 feet, incapable of effective
work at 10,000 feet. On the other hand a subject may become
adapted to a limited degree of oxygen deprivation.

The qualitative changes in behavior have also been reported
in some detail. Depletion of the oxygen supply is accompanied
by muscular twitching and tremor, incoérdination, and over dis-
charge. The subject becomes less and less capable of normal
attention, until finally he will heed instructions only when they are
given insistently and in a loud voice. In the final stages emotional
disturbances are not uncommon. Some subjects express uncon-
trolled anger, some become sullen, and others become silly.

SEASONAL VARIATIONS

Peaks (70) has reported a rather elaborate historical and experi-
mental study of seasonal variations in efficiency. His own experi-
ments, like those of earlier investigators, show that there are
seasonal variations in mental efficiency. In this case there was a
depression in mental ability which appeared in the winter shortly
after the appearance of a more noticeable and longer lasting depres-
sion in physical capacity. The interpretation of this fact is, how-
ever, almost impossible. Peaks seems to think that the larger
yearly variations in abilities are due to the combined influence of
temperature and light strength. But he admits that the following
MENTAL WORK 477

factors may have some influence: food, clothing, body temperature,
humidity, air pressure, practice effects, previous mental and physical
development, health, etc. If this is true, correlations between
efficiency and some of the simpler and more definable conditions
of life are probably more worth noting than correlations between
efficiency and such vague conditions as are covered by the names
of the months and the seasons.

Diurna.L VARIATIONS

Since Marsh in 1906 published his study of the diurnal course of
efficiency a number of attempts have been made to test and extend
his findings. These investigations have usually been concerned
with diurnal variations either as indicators of the amount of fatigue
induced by the work of school or factory, or as indicators of varia-
tions in efficiency. Heck (37) and Martyn (58, 59), for example,
seem to have been interested mainly in the effect of the day’s
work upon school children, while Winch (93, 94, 95) and Gates (35,
36) came at the matter more with the purpose of determining the
most.efficient hours of the school child’s day. These differences in
point of view, however, are not, in recent writers at least, accom-
panied by a narrowness in the interpretation of the facts of diurnal
variation. Those who are interested in such variations as indicators
of fatigue would hardly deny the complexity of a valid conception
of mental fatigue, and those who are interested in these variations
as indicators of an organic periodicity would certainly admit their
dependence upon the type of occupations indulged in during the
day.

From the facts now available it is impossible to arrive at any
statement of the diurnal course of efficiency which has a very wide
applicability. According to Gates, Peaks, and Robinson (76) men-
tal efficiency of school children rises during the morning, shows a
low point early in the afternoon and after that another rise which
probably is followed by another low point. Such findings are not
to be criticized as facts and as facts they probably may be used to
advantage under certain peculiar conditions. Nevertheless they
can not be considered as broadly typical of diurnal fluctuations.
Hollingworth (39) found a falling off in mental efficiency during the
day and so did Arai (3). Muscio (65) as has been reported, found
that feelings of fatigue increase throughout the day except for an
early afternoon recuperation. One can explain these differences in
terms of the subjects employed, the nature of the day’s work, changes
478 E. S. ROBINSON

in temperature and other atmospheric conditions, meals and the like.
If this is true, it seems reasonable to believe that, insofar as there
are differences in the efficiency of work at different hours, these
differences are themselves to be explained in terms of the subjects,
the work, etc., rather than in terms of the characteristic natures of
different times of day. Each time of day, like each time of year,
covers such a host of psychological possibilities that it seems futile
to work out its correlates except for some extremely practical and
immediate purpose. It may, of course, be well worth while for
employers and educators to know the specific diurnal variations of
specific groups of employees or students working under conditions
which for them are standard.

INDIVIDUAL AND Group DIFFERENCES

The literature on mental tests contains a wealth of material
regarding differences in mental efficiency between individuals and
groups of individuals, a review of which would hardly be appropriate
in this place. Investigators concerned with topics usually included
under mental work have not as a rule paid much attention to indi-
vidual or group differences except insofar as these differences have
had a bearing on the validity of experimental results. Once in a
while an exception has appeared. For instance, Winch (94) pre-
sents a relationship between the nature of diurnal variations and
age; and Garth (33) presents a possible relationship between
fatiguability and race. In another study (34) Garth attacks the
claim made by Meumann, Kraepelin, and others that work curves
in computation experiments fall into types. Garth admits that the
curves collected by him vary within wide limits as to general trend,
but none save the average curve is modal.

REFERENCES

1. Air Service Medical Manual, War Department—Air Service Division of Military
Aeronautics: Washington, D. C., 1918. Pp. 446.

2. Atuport, F. H. The Influence of the Group upon Association and Thought.
J. of Exper. Psychol., 1920, 3, 159-182.

3. Arar, T. Mental Fatigue. ‘Teachers’ College Contributions to Education, 1912,
No. 54. Pp. 115.

4. Arurrt, A. H. The Effect of Alcohol on the Intelligent Behavior of the White
Rat and its Progeny. Psychol. Monog., 1919, No. 115. Pp. 50.

5. Arps, G. F. Work with Knowledge of Results versus Work without Knowledge
of Results. Psychol. Monog., 1920, No. 125. Pp. 41.

6. Asn, I.E. Fatigue and Its Effects upon Control. Arch. of Psychol., 1914, No. 31.

7. Bass. Experiment in School Room Ventilation with Reduced Air Supply through
Individual Ducts. Trans. Amer. Soc. Heat. t& Vent. Eng., 1913, 19, 328. (Re-
viewed by Stecher.)
MENTAL WORK 479

8. Bass. The Recirculating of Air in a Schoolroom in Minneapolis. Trans. Amer.

9.
10,

Il.
12.

13.

14.

15.

16.

17.

18.

19.

20.

21.

22.

23.

24.
25.

26.

27.

28.

29.

30.

31.

32.

Soc. Heat. &% Vent. Eng., 1915, 21, 109. (Reviewed by Stecher.)

Bacay, E. Effects of Oxygen Deprivation. J. of Comp. Psychol., 1921, 1, 97-113.

Beruner, A. The Influence of Mental Work on the Visual Memory Image.
Amer. J. of Psychol., 1918, 29, 355-370.

Berry, C.S. Effects of Smoking on Adding. Psycnou. Butt., 1917, 14, 25-28.

Burnoam, W.H. The Effect of Tobacco on Mental Efficiency. Ped. Sem., 1917,
24, 297-317.

Burtt, H. E. The Effect of Uniform and Non-Uniform Illumination upon
Reaction Times, with Special Reference to Street Illumination. J. of Exper.
Psychol., 1916, 1, 155-182.

Buss, A. D. Tobacco Smoking and Mental Efficiency. N.Y. Med. Jour., 1914,
99, 519-527.

CasseL, E. E. & Dattenpacn, K. M. The Effect of Auditory Distraction upon
the Sensory Reaction. Amer. J. of Psychol., 1918, 29, 129-143.

Cuarman, J.C. A Study of Initial Spurt in The Case of Addition. J. of Educ.
Psychol., 1915, 6, 419-426.

Cuapman, J. C. The Effect of Rapid Changes of Work on Performance. J. of
Exper. Psychol., 1917, 2, 165-170.

Cuapman, J. C. & Fever, R. B. The Effect of External Incentives on Improve-
ment. J. of Educ. Psychol., 1917, 8, 469-474.

Cuapman, J. C. & Notan, W. J. Initial Spurt in a Simple Mental Function.
Amer. J. of Psychol., 1916, 27, 256-260.

Docxeray, F.C. The Effects of Physical Fatigue on Mental Efficiency. Kansas
Univ. Sci. Bull., 1915, 9, 197-243-

Docxeray, F. C. & Isaacs, S. Psychobiological Research in Aviation. J. of
Comp. Psychol., 1921, 1, 115-148.

Dovcgz, R. The Refractory Phase of the Protective Wink Reflex. Amer. J. of
Psychol., 1913, 24, 1-7.

Dopce, R. Mental Work; A Study in Psychodynamics. Psychol. Reo., 1913, 20,
1-42.

Donce, R. The Laws of Relative Fatigue. Psychol. Rev., 1917, 24, 89-113.

Doncg, R. & Benepvict, F. G. Psychological Effects of Alcohol. Washington,
D. C., Carnegie Institution, 1915. Pp. 281.

Dones, R. & Benepict, F.G. Psychology: Neuro-muscular Effects of Moderate
Doses of Alcohol. Proc. Nat. Acad. Sci., 1915, 1, 605-608.

Duntar, K. Effects of Alcohol on Hand and Eye Coérdination. Psychobiol.,
1920, 2, 187-191.

Evans, J.E. The Effect of Distraction on Reaction Time, with Especial Reference
to Practice and the Transfer of Training. Arch. of Psychol., 1916, No. 37,
Pp. 106.

FEeRNBERGER, §. W. Influence of Mental and Physical Work on the Formation
of Judgments in Lifted Weight Experiments. J. of Exper. Psychol., 1916, 1,

08-532.

Noe W. Arbeitsversuche an der Schriebmaschine. Psychol. Arbeit.,
1912, 6, 419-451.

Frozserc, $. Effects of Smoking on Mental and Motor Efficiency. J. of Exper.
Psychol., 1920, 3, 334-346.

Gartu, T. R. Mental Fatigue. Arch. of Psychol., 1918, No. 26. Pp. 85.
480 E. S. ROBINSON

33-

34.
35)

36.
37:

38.

39-
40.
4I.
42.
43-
44.

45.

47-
48.

49.

50.

SI.

52.

53-
54.
55-
56.

57-

58.

Gartu, T. R. Racial Differences in Mental Fatigue. J. of Appl. Psychol.
1920, 4, 235-244.

Gartu, T. R. Work Curves. J. of Educ. Psychol., 1919, 10, 277-283. ;

Gates, A. I. Diurnal Variations in Memory and Association. Univ. of Cal.
Publ. in Psychol., 1916, 1, 323-344.
Gates, A. I. Variations in Efficiency During the Day, etc. Univ. of Cal. Publ.
in Psychol., 1916, 2, 1-156. .
Hecx, W. H. The Efficiency of Grammar Grade Pupils in Reasoning Tests in
Arithmetic at Different Periods of the School Day. Lynchburg, Va.: Bell, 1913.

Houuincworts, H.L. The Influence of Caffeine on Mental and Motor Efficiency.
Arch. of Psychol., 1912, No. 22. Pp. 167.

Houuincworta, H. L. Variations in Efficiency during the Working Day.
Psychol. Rev., 1914, 21, 473-491.

Ho.unceworts, H. L. & PorFENBERGER, A. T., Jr. Applied Psychology. N. Y.:
Appleton, 1917.

loteyxo, I. Le defense psychiques. Rev. Psychol., 1913, 6, 3-22; 147-155-

Jounson, H.M. Dynamogenic Influence of Light. Psychobiol., 1920, 2, 351-374-

Jounson, H. M. & Pasar, F.C. Psychological Effects of Deprivation of Oxygen,
etc. Psychobiol., 1920, 2, 193-236.

Jounson, H. M. Research in Aviation during the year 1919. Science, 1920, 51,
449-452.

Jounson, O. J. Effects of Smoking on Motor and Mental Efficiency. Psychol.
Clinic, 1918, 12, 132-141; 230-235.

. Jones, E.D. The Administration of Industrial Enterprises. New York: Long-

mans, Green, 1919. Pp. 442.

Jorcer, J. B. Ueber Associationen bei Alkoholikern. Monatsch. f. Psychiat. u.
Neurol., 1915, 37, 246-266; 323-332.

Karuson, K. J. The Mental Processes as Affected by Alcohol. Ped. Sem., 1913,
20, 322-342. (A review of literature on alcohol.)

Lance, J. & Sprecut, W. Neue Untersuchung iiber die Beinflussung der Sinnes-
funkionen durch geringe Alkoholmengen. Zsch. f. Pathopsychol., 1915, 3,
155-256.

Lancretp, H. S. On the Psycho-physiology of a Prolonged Fast. Psychol.
Monog., 1913-1914, No. 71. Pp. 62.

Lancretp, H. S. Suppression with Negative Instruction. Psychol. Rev., 1911,
18, 411-424.

Lasutey, K.S. The Effects of Strychnine and Caffeine upon the Rate of Learning.
Psychobiol., 1917, 1, 141-171.

Lasutey, K. S. The Effect of Strychnine upon Habit Formation. Psycuou.
BuLt., 1916, 23, 79-80.

Leuman, A. Ueber den Stoffwechsel Wahrend Geistiger Arbeit. Bericht V.
Congress f. Exper. Psychol., 1912, 136 ff.

Macur, D. I. & Isaacs, $. Action of Some Opium Alkaloids on the Psycho-
logical Reaction Time. Psychobiol., 1917, 1, 19-31.

Macut, D. I., Isaacs, S., & GREENBERG, J. Action of Some Anti-phyretic
Analgesics on the Psychological Reaction Time. Psychobiol., 1918, 1, 327-338.

Marsu, H. D. Individual and Sex Differences Brought out by F asting. Psychol.
Rev., 1916, 23, 437-445.

Martyn, G. W. The Evidence of Mental Fatigue during School Hours. J. of
Exper. Ped., 1911, 1, 39-45, 137-141.
59-

60.
61.
62.
63.
64.
65.
66.
67.

68.

MENTAL WORK 481

Martyn, G. W. The Evidence of Mental Fatigue in the Speed and Accuracy of
Mental Work before and after the School Day. J. of Exper. Ped., 1911, 1,
142-147.

McDovaeatt, W. & Smitn, M. British Med. Research Council. Special Report,
Series No. 56, 1920, 1-34. :

Mutzs, W. R. Effects of Alcohol on Psychophysical Functions. Washington,
D.C.: Carnegie Institution, 1918. Pp. 144.

Mies, W. R. Some Psycho-physiological Processes as Affected by Alcohol.
Proc. Nat. Acad. of Sci., 1916, 2, 703-709.

Moreay, J. J. B. The Effect of Fatigue on Attention. J. of Exper. Psychol.,
1920, 3, 319-333.

Morean, J. J.B. The Overcoming of Distractions and Other Resistances. Arch.
of Psychol., 1916, No. 35. Pp. 84.

Muscio, B. Feeling Tone in Industry. Brit. J. of Psychol., 1921, 12, 150-162.

Muscio, B. Is a Fatigue Test Possible? British J. of Psychol., 1921, 12, 31-46.

Parnter, W. S. Efficiency in Mental Multiplication under Extreme Fatigue.
J. of Educ. Psychol., 1916, 7, 25-30.

Patrizi, M. L. La simultanéité (Dynamogenie et Inhibition) entre le travail
mental et le travail musculaire volontaire unilateral ou symétrique. Arch.
ital. de biol., 1912, 57, 170-186.

69. Patrizi, M. L. Nouveau “test” mental: un indice autographic et inconscien

70.

71.

72.
73-
74.
75.
76.
77-
78.
79-
80.
81.

82.

du pouvoir inhibiteur. Arch. ital. de biol., 1912, 57, 190-204.

Peaks, A. Z. Periodic Variations in Efficiency. Educ. Psychol. Monog., 1921.
No. 23. Pp. 95.

Puiturrs, F. M. A Comparison of the Work done in Successive Minutes of a Ten-
Minute Practice Period in the Fundamentals of Arithmetic. J. of Educ.
Psychol., 1916, 7, 271-277.

PorrenBERGER, A. T., Jr. The Effects of Strychnine on Mental and Motor
Efficiency. Amer. J. of Psychol., 1914, 25, 82-120.

Porrensercer, A. T., Jr. & Tatuman, G.G. Variability in Performance during
Brief Periods of Work. Psychol. Rev., 1915, 22, 371-376.

Pressey, S. L. The Influence of Color upon Mental and Motor Efficiency.
Amer. J. of Psychol., 1921, 32, 326-356.

Rossins, H., Smita, D. & Wasusurn, M. F. The Influence of Fatigue on
Affective Sensitiveness to Colors. Amer. J. of Psychol., 1915, 26, 291-292.
Rostnson, L. A. Mental Fatigue and School Efficiency. Bull. Winthrop Normal

&F Industrial College of S. C., 1912, 5, No. 2. Pp. 56.

Scuituinc, W. The Effect of Caffeine and Acetanelid on Simple Reaction Times.
Psychol. Rev., 1921, 28, 72-79.

Smita, M. A Contribution to the Study of Fatigue. Brit. J. of Psychol., 1915-
17, 8, 327-350.

Srarcu, D. & Asn, I. E. The Mental Work Curve. Psychol. Rev., 1917, 24>
391-402.

Srecuer, L. I. The Effect of Humidity on Nervousness and General Efficiencyt
Arch. of Psychol., 1916, No. 38. Pp. 94.

Trornpike, E. L. Educational Psychology. Teachers College, Columbia iWintvs
N. Y., 1914, Vol. 3, pp. I-140.

Taoeaniee, E. L. Fatigue in a Complex Function. Psychol. Rev., 1914, 21,
402-407.
482 E. S. ROBINSON

83.
84.
85.
86.
87.
88.
89.
go.
gl.
92.

93-

94.

95-

m Om & & OD w ®&

™

™~

Tuornpixz, E. L. Notes on Practice, Improvability and the Curve of Work,
Amer. J. of Psychol., 1916, 27, 550-565.

Tuornpixs, E. L. The Curve of Work and the Curve of Satisfyingness. J. of
Appl. Psychol., 1917, 1, 265-7.

Tuornpike, E. L. The Curve of Work. Psychol. Rev., 1912, 19, 165-194.

Tuornpixe, E. L., Ruger, H. A. & McCat, W. A. The Effects of Outside Air
and Recirculated Air upon the Intellectual Achievement and Improvement of
School Pupils. School &F Society, 1916, 3, 679.

Tuornvike, E. L. & Oruers. Ventilation in Relation to Mental Work. Teachers
Coll., Columbia Unio., N. Y., Contributions to Education, 1916.

Topp, J. W. Reaction to Multiple Stimuli. Arch. of Psychol., 1912, No. 25.
Pp. 65.

Toiman, E. C. Retroactive Inhibition as Affected by Conditions of Learning.
Psychol. Monog., 1917-18, No. 107. Pp. 50.

Ursan, F. M. Prof. Dodge’s recent Discussion of Mental Work. Amer. J. of
Psychol., 1913, 24, 270-274.

Vernon, H. M. Causation of Industrial Accidents. Ministry of Munitions,
Health of Munition Workers, Com. Mem. 1918, No. 21. Pp. 46.

Watson, J. B. Psychology from the Standpoint of a Behaviorist. Phila.: Lippin-
cott, 1919. Pp. 348-392.

Wincn, W. H. Mental Adaptation during the School Day as Measured by
Arithmetical Reasoning. 2 parts. Jour. of Educ. Psychol., 1913, 4, 17-28
71-84.

Wincu, W. H. Mental Fatigue in Day School Children as Measured by Arith-
metical Reasoning. Brit. Jour. of Psychol., 1911, 4, 315-341.

Wincn, W. H. Mental Fatigue in Day School Children as Measured by Imme-
diate Memory. J. of Educ. Psychol., 1912, 3, 18-28; 75-82.

. ZonpEK, B. Die ermiidende Einfluss von rein psychischer, affectloser Arbeit auf

den peripheren Teil der Muskelarbeit. Arch. f. Anat. u. Physiol., 1916, 360-366.
In addition to the above, the following reviews were consulted:

. Docxeray, F. C Fatigue, Work and Inhibition. Psycnox. BuLL., 1920, 17

322-330.

. Horurncwortu, H. L. Psychological Aspects of Drug Action. Psycuou. Buut.,

1912, 9, 420-423.
PorFenBERGER, A. T., Jr. Psychological Effects of Drugs. Psycuou. BuuL.,
1914, 11, 418-421.

. PorFENBERGER, A. T., JR. Psychological Effects of Drugs. Psycuou. BuLt.,

1916, 13, 434-436.

. Porrenspercer, A. T., Jr. Psychological Effects of Drugs. Psycuou. BuLL.,

1917, 14, 409-411.

. PorrenBercer, A. T., Jr. Drugs. Psycuou. Buu.., 1919, 16, 291-292.
. Strone, E. K., Jr. Fatigue, Work, and Inhibition. Psycuox. Buut., 1913, 10,

440-450.

. Srrone, E. K., Jr. Fatigue, Work, and Inhibition. Psycuox. BuLu., 1914, 115

412-417.

Strone, E. K., Jr. Fatigue, Work, and Inhibition. Psycuot. BuLL., 1915, 12,
416-419.

Strone, E. K., Jr. Fatigue, Work, and Inhibition. Psycuot. Butt., 1916, 13,
430-433.

. Wetts, F.L. Fatigue. Psycuou. Buuu., 1912, 9, 416-420.
RECENT LITERATURE ON THE PSYCHOLOGY
OF THE MUSICIAN

BY MAX SCHOEN

Carnegie Institute of Technology

The most significant contributions to the psychology of the
musical mind as it functions in the musically talented person, or
artist, have appeared since the year 1915. Prior to that year there
was an abundance of psychological literature concerning the nature
of the material of music, from the single tone to the complete
melody, but the creator and reproducer of music received only a
passing interest and an occasional expression of opinion from
psychologists. Stumpf (17), for instance came in contact with a
Spanish boy prodigy and disposed of him in a two-page paper!

The musician began to come into his own in experimental
psychology with Rupp’s (14) contribution to the psychology of
talent in the form of a prospectus for future research. In 1913,
Revesz (11) made a preliminary report on an analysis of a boy
especially gifted musically, but it was not until 1916 (12) that the
full report appeared, and not until 1920 (13) that he published an
inventory of musicalness based on his study of the boy prodigy.
The problems with which Revesz concerns himself in the 1920 study
are: 1. Can a method be devised which will identify the musical
mind before the beginning of instruction, even in cases less excep-
tional than that of the boy examined; and, 2. Could such a method
be applied to adults and to the musically trained? His aim, then,
was to find a method of identifying the musical, adapted to children
and adults, and to the musically trained and untrained. For this
purpose he devised eight tests, based upon musical and acoustical
factors, and gave these to 63 boys and girls from seven to twelve
years of age. The tests are as follows: 1. The rhythmic sense.
Twelve rhythmic groups were arranged in order of difficulty. In
the first part of the test the rhythms were presented monotonically
and the children were asked to reproduce the rhythm with hand-
claps. In the second part of the test the same rhythmic forms
were presented melodically and a like response was called for from
the children. 2. Absolute pitch. For the second test the experi-

483
484 MAX SCHOEN

menter sounded a tone on the piano and the observers were called
upon to reproduce the tone on the instrument. 3. Octave reco gnition
and transposition. This test Revesz found too easy for his obser-
vers. 4. Relative pitch. He first produced an interval and then
sounded any tone on the piano and asked the child to reproduce
vocally the given interval, with the new tone for its basis. 5 and
6. The harmonic sense. In these tests the children were asked to
reproduce vocally the constituent tones of chords. 7and8. M, elodic
memory and playing by ear. The seventh test deals with memory
for melody, and the eighth with the ability to play by ear a known
tune.

Revesz assumed the vocal reproduction of a melody to have the
greatest diagnostic value, and by correlating the other tests with the
seventh test, he concluded that the sense of rhythm does not possess
great musical significance, but that the spontaneous instrumental
reproduction of a known melody, absolute pitch, vocal transposition
of an interval, and also the ability to analyze simultaneous intervals,
are characteristic manifestations of musicalness. Revesz’s investi-
gation has two great merits—his inventory has an empirical basis,
and his test material has musical significance. ‘The number of his
observers was too small, however, for the establishment of norms.

In his study on the earliest appearance of talent (11), Revesz
claims that musical talent necessarily makes its appearance in
early youth; at first appears technical equipment, or virtuosity,
later, power of interpretation, and then creative power. An exam-
ination of the great composers shows that the full development of
the creative powers is reached between the twentieth and thirtieth
years.

In contrast to Revesz, Seashore’s inventory (15) has a theoretical
basis, while the main factors of his inventory are acoustical rather
than musical. One is entirely justified in offering the criticism
that Seashore measures individual differences in acoustical sensory
capacities, rather than differences in musical powers. The Seashore
tests are rapidly gaining admittance into the class-room and studio
as a basis for vocational guidance in music and for this reason, if
for no other, they should be subjected to a rigid and thorough
examination. Seashore’s work is epoch-making in that he is the
first to devise rigid scientific tests for measurement, the first to
establish norms for musical talent, and the first to make the tests
available for the music teacher in studio and class-room. No
further mention of this first treatise on musical talent is necessary
as it is readily available to readers in English.
PSYCHOLOGY OF THE MUSICIAN 485

Several briefer studies are available that help to throw light
upon specific factors in the more extensive treatments of Revesz
and Seashore. Schiissler (18) attempts to account for the unmusical
child in a paper based upon historical and statistical material. In
the first part of the paper, he reviews briefly the works of Billroth,
Max Meyer, Stumpf, Von Meyer, Rupp, Revesz, and several others,
as well as the contributions on the anatomical and physiological
bases of musicalness. His own contribution consists of a statistical
examination of two hundred unmusical children, the criterion for
unmusicalness being attainment of a certain grade in school singing.
Schissler’s conclusions, based upon the historical review and his
statistical data, are as follow: 1. There are between five and ten
per cent. of unmusical people. The majority of the unmusical, if
not all, are not beyond profiting by musical instruction, and should
therefore not be neglected, but rather given well-planned and in-
tensive instruction. 2. For musical or unmusical capacity, interval
memory is of greatest significance. 3. Absolute pitch is of signifi-
cance only for outstanding musical gift. The unmusical, as com-
pared with the musical, are as a whole less gifted than the musical,
i.e., they are not only inferior musically, but also in other mental
capacities. 4. Forty-one per cent. of the unmusical, fifty-seven per
cent. of the half-musical, and seventy-nine per cent. of the musical,
make satisfactory progress in school. The school work of the musical
is fifteen per cent. superior to that of the unmusical.

Schiissler’s findings on the relation between intelligence and
musicalness are in accord with those of Stumpf, who found that
musical children are, as a whole, better students than the unmusical.

The Pannenborgs (10) examined 3,860 children, 423 musicians,
and the biographies of twenty-one composers in an attempt to
derive the intellectual and emotional characteristics of the musician.
Their conclusions agree with those of Schissler and Stumpf. Of
all the children, 342 boys and 152 girls were designated as musical.
The musical children were advanced in their school work beyond
their years, while the musician was found to have extensive intel-
lectual interests and to be widely gifted, particularly in languages,
literature, history, geography, mathematics, natural history, and
drawing.

Bernfeld (x) claims that “the individual’s reaction to music is
by no means wholly decided by the nature and quantity of his
psychophysical tendencies. It is influenced to a certain degree by
the will to be or not to be musical. Even when accurate tests
486 MAX SCHOEN

have shown that a person possesses all the elements of musical
ability, it cannot be foretold with certainty whether he can acquire
musical appreciation, for it is possible that an inhibition with retro-
active force against music may exist in him, a will to be unmusical,
or at least to seem so to himself and others.”

Kovacs (6) contributes a much needed study on musical memory.
His own experience as a piano teacher taught him that his pupils,
although very musical, were incapable of playing through a musical
composition from memory with any degree of certainty, in spite of
great effort. In planning his experiment, the author concluded
that the visual factor did not play an important rdéle in playing
from memory, since the pupils could write the composition from
memory and could also easily memorize poetry, numbers, etc.
Neither could he ascribe the difficulty to motor adjustments,
because of the technical proficiency of the pupils, and because the
memory failures did not occur at the most difficult passages. The
pupils had studied the piano at least eight years, so that their
deficiency could not be due to lack of visual and motor codrdination.
He therefore concluded that the difficulty must be of an acoustical
nature. In other words, the pupils could not hear, inwardly, that
which they were playing. He asked the performers to play the
melody of a composition with which they were thoroughly familiar
an octave above and found that they were incapable of so doing.
Neither could they reproduce vocally the alto or bass voices of a
memorized fugue. With these problems before him, he initiated
an investigation to answer the following question: How is a musical
composition most readily committed to memory, by practice at
the piano, or away from the piano? His observers were all accom-
plished pianists and piano teachers. He concludes that the diffi-
culty that music students experience in memorizing musical com-
positions is due to the lack of a thoroughly trained “inner” and
“outer” tonal hearing. By “outer” hearing he means the ability
to name tones heard, and by “inner” hearing, the ability to image
tones from the staff. The student should begin his piano study,
according to this writer, with note-reading.

In another study (7), Kovacs deals with the relation between
recognition and reproduction as they function in musical memory.
Is recognition necessarily involved in reproduction or are the two
independent factors or drifts of memory? Three differences were
found to exist between the two types of memory: 1. The attitude
in recognition is more passive, in reproduction, more active; 2. In
PSYCHOLOGY OF THE MUSICIAN 487

recognition attention is fixed more on the memory material as a
whole, in reproduction on the sequences of the constituent sections;
3. The imagery used in recognition is often different from that used
in reproduction. ‘The author concludes that he who possesses good
recognition, but poor reproduction, may become a good listener,
but not a performer. The artist must possess both, while the
teacher stands more in need of recognition, and the virtuoso of
‘reproduction. The student, in preparing for public performance,
should be careful not to confuse the two types of memory in his
practice. Reproduction must predominate over mere recognition.

Heinitz’s study (4) on musical reproduction is most valuable in
that he suggest, in a preliminary way, ways and means to determine
native adaptibility for musical performance on different instruments.
The instruments used by the author were the piano, violin, and
flute.

Gatewood’s research (3) on finger reactions is suggestive in its
bearing on musical performance, and therefore valuable for voca-
tional guidance in music, in that she finds that there are some indi-
viduals who, in spite of unlimited practice, will not attain the speed
and accuracy which some others show at the initial trial. Morton
(9) reports the results of discontinued experiments on the degree of
accuracy attained by pianists in playing double against triple time.

Copp (2) argues that although musical ability is innate and
inherited, “almost everyone possesses the heredity . . . although
great genii will doubtless continue to be sporadic and unaccountable,
real musical ability is much more common than has been supposed

. the very large majority of children lose their native musical
ability through lack of training of the ear and mind during their
most susceptible period.” Copp bases this sweeping statement on
the fact that eighty per cent. of normal children can be trained,
according to her experience, to sing “‘Center C” on demand, and to
recognize it when it is played or sung. “This surely indicates that
musical talent is much more widespread than has been thought.”

The study of the nature of the musical response still follows in
the main the questionnaire, introspective, and purely speculative
methods. But one brief recent study of the physiological effects
of music is available, and that one is incomplete. Hyde and
Scalapino (5) experimented with Tschaikowsky’s Death Symphony
(which symphony is that?), the Toreador song from Carmen, and
a Sousa march, in order to “ascertain the effect of different kinds
of music upon the heart and blood pressure in individuals who are
488 MAX SCHOEN

known to have musical talent and are fond of music; also persons
who are indifferent and have no fondness for music, in neurasthenics
and in some animals.” They find that the minor tones of the
symphony increase the pulse rate and action current of the ventric-
ular contraction, and lower the systolic and diastolic pressures.
The Carmen song and the march increased the systolic and pulse
pressure, but the former also increased the pulse rate, with decreased
diastolic pressure and action current, while the march slowed the
cardiac cycle and increased its action currents. The authors do
not state whether these results were obtained with the musical or
unmusical subjects.

Washburn (19) seeks to answer the questions: 1. How far does
musical enjoyment or pleasure depend on principles not governing
our other enjoyments? 2. How far and why is the relation of
music to emotion unique? For the first question the sources of
pleasure derived from single tones and tonal combinations, simul-
taneous and in sequence, are considered. The author concludes
that all sources of musical pleasure are sources not peculiar to
music. We enjoy music because we enjoy other things. The
uniqueness of music in relation to the emotions is that it reaches
these directly, without reference to associations based upon the
individual’s experience. ‘Musical sound is the kind of sound that
is produced under stress of emotion; rhythmic sound means that
the emotion is magnified and exalted by being shared with the
whole mass of one’s fellow beings.”

To Lee (8) the question of what music is is synonymous with
what music does to the mind of the hearer, or ‘more correctly, of
what the mind of the hearer does in response to the music which
he hears.” With this as a guiding principle, the writer carried out
an experiment in order to determine the varieties of musical experi-
ience. The subjects were asked in a questionnaire to react to the
question: When music interests you at all, has it for you a meaning
which seems beyond itself a message; or does it remain just music?
The questionnaire contained also sets of queries to determine the
degree of musical endowment of the subjects. Half of the subjects
answered that to them music imparted a message, a number of
these reporting that they were attracted to music primarily because
of the images, emotions, and trains of thought which it suggested.
The other half contended that whereas they found a meaning in
music, that meaning was inseparable from the music itself, and
that visual images or emotional suggestions were either excluded
PSYCHOLOGY OF THE MUSICIAN 489

or insignificant. The subjects who repudiated the message of
music, to whom music had a meaning in itself, were found to be
those who were most musical.

REFERENCES

. BernFrerp, S. Zur Psychologie der Unmusikalischen. Arch. f. d. ges. Psychol.,

1915, 34.

2. Copp, E. F. Musical Ability. J. of Hered., 1916, 7, 297-304.

- Gatewoop, E. L. Individual Differences in Finger Reaction. Psychol. Monog.,

1920, No. 126.

. Heitz, W. Experimentelle Untersuchungen iiber musikalische Reproduktion.

Arch. f. d. ges. Psychol., 1915, 34.

. Hype, I. H. & Scatapino, W. The Effect of Music upon Electrocardiograms and

Blood Pressure. Amer. J. of Physiol., 1918, 46, 35-38.

. Kovacs, 8. Untersuchungen tiber das musikalische Gedachtnis. Zsch. f. angew.

Psychol., 1916, 11.

. Kovacs, S. Uber das Verhialtnis der erkennenden und mitteilenden Gedachtnisse

auf musikalischem Gebiet. Arch. f. d. ges. Psychol., 1918, 37-

. Lez, V. Varieties of Musical Experience. North Amer. Rev., 1918, 207, 748-757.

9. Morton, W. B. Some measurements of the accuracy of the time-intervals in

10.
It.

12.

13.
14.

15.
16.
17.

18.
19.

playing a keyed instrument. Brii. J. of Psychol., 1920, 10, 194-198.

Pannensore, H. J. & W. A. Die Psychologie des Musikers. Zsch. f. Psychol.,
1915, 73, 91-136.

Révész, G. Erwin Nyiregyhazi. Psychologische Analyse eines musikalischen
hervorragenden Kindes. Leipzig, 1916, Veit u. Co.

Révész, G. Ueber das fruhzeitige Auftreten der Begabung. Zsch. f. angew,
Psychol., 1919, 15.

Révész, G. Prifung der Musikalitat. Zsch. f. Psychol., 1920, 85, 163-209.

Rupr, H. Uber die Prifung musikalischer Fahigkeiten. Zsch. f. angew. Psychol.,
1914, 9, 1-76.

Szasuore, C. E. The Psychology of Musical Talent. Boston, Silver, Burdett,
1919. Pp. xvi + 288.

Srasuore, C. E. Vocational Guidance in Music. (Univ. of Iowa Monog., 1916,
8. 1, N. 2.) ;

Srumpr, C. Akustische Versuche mit Pepito Arreola. Zsch. f. angew. Psychol.,
1908, 2.

Scnissuer, H. Das unmusikalische Kind. Zsch. f. angew. Psychol., 1916, 11.

Wasusurn, M. F. Psychology of esthetic experience in music. N. E. A. 1916.
THE PSYCHOLOGY OF LANGUAGE
BY ERWIN A. ESPER

Ohio State University

The present review deals with the contributions to the psychol-
ogy of language which have appeared in periodical literature during
the five-year period, 1917-1921. Such a review, incomplete though
it may be, is yet sufficient to indicate that this field is not being
extensively worked by psychologists, and this in spite of the in-
creasing recognition of the fact that a large proportion of the
apparently most perplexing problems in psychology are really
language problems. Indeed, it seems likely that many of the
problems of the other sciences will find their solution in an investi-
gation of the genesis of the verbal reactions involved in the state-
ment of these problems. The linguists and the philologists, too,
have long been awaiting an adequate psychology for the interpre-
tation of their data; those who have found the traditional psycho-
logical interpretations no longer satisfactory have in some cases
undertaken to work out for themselves theories as to the nature
of language behavior. From all of these points of view it seems
highly important that language receive more attention from those
whose proper business it is to investigate the fundamental princi-
ples underlying human behavior. Indications are not lacking that
the next few years will witness an increasing amount of work by
psychologists in this field.

The problem of the nature of language has received most
attention from those engaged in the study of speech disturbances.
Barat (1) summarizes the leading theories which have been thus
evolved. He regards language as dependent upon the development
of a system of automatic associations. which function in the co-
ordination of speech movements, in the analysis and interpretation
of sensation, and in the evocation and identification of images.
These associations cannot develop without a parallel differentiation
of certain regions of the cerebral cortex, whose integrity therefore
becomes necessary to the automatism of speech.

Sechehaye (17) has written an exposition and interpretation of
the linguistic doctrines of the philologist Ferdinand de Saussure.

490
THE PSYCHOLOGY OF LANGUAGE 491

According to these doctrines, we must distinguish between the
speech acts of individuals (/a parole) and the system of conventions
existing in the speech habits of a community (la langue). The
latter is a system of arbitrary signs whose values are not inherent
but wholly conventional. The fundamental principle is that there
be as many distinct signs as there are distinct ideas. Ideas, how-
ever, do not precede their signs, since a community, not possessing
a joint mind, can never seize an idea before assigning to it a sign.
Both signs (i.¢., significant sound sequences) and the ideas which
they represent have their individual values by virtue of the fact
that each is distinctive as compared with the others. There is thus
at any given time an equilibrium between all the elements of
language. Because of its arbitrary, conventional character, lan-
guage is not influenced in its organization by the mind, but is very
susceptible to the influences of other factors, such as phonetic
change. When influences of the latter sort disturb the equilibrium
of elements, a new equilibrium is brought about by interpretation,
1.é.,a disturbance of one element is made good by the adaptation of
others. Thus, “language is at any moment a fortuitous resultant
of this arbitrary combination of the sounds and the ideas which
constitute it.”

Schiefferdecker (15) has thrown some interesting light on the
problem of the origin of speech. His investigation of the phylo-
genetic evolution of the muscles of mastication shows a marked
differentiation in man as compared with animals, a differentiation
due probably to the functioning of these muscles in speech. There
is no gradual evolution through the primates to man, and in man
himself the differentiation begins relatively late in the embryonic
period and is not complete until the twelfth year of life. This
indicates a late origin of speech in man’s racial history, an origin
which was contingent upon an adequate development of the brain.
Of the three muscles investigated (masseter, pterygoideus internus,
temporalis), the masseter showed by far the greatest differentiation,
and would hence appear to be especially important in speech.

Thumb and Marbe in 1go1 reported the results of an investiga-
tion of the psychological basis of analogical changes in language.
From these results they concluded that the effectiveness of a word-
association in bringing about analogical changes increases with the
number of members of a speech community in whom it is present
and with a decrease in the average time required for the association
to take place. Esper (6) repeated this investigation with 126
492 ERWIN A. ESPER

American subjects (100 educated adults, 11 children, 15 uneducated
adults). The results were similar to those of the German investiga-
tion; in both cases it was found that the greater the number of
subjects who respond with an identical word, the smaller is the
average association time for this response; in both languages, words
of a given grammatical category were found to be associated pre-
dominantly with words of the same category; numerals from one to
ten were associated predominantly with the next higher numeral;
and adjectives with adjectives of opposed meaning.

Barnils (2) points out that heredity in language constitutes a
fundamental problem still without solution. He believes that the
cases reported of individuals producing speech sounds which have
not occurred in their environment but which were characteristic of
their ancestors call for a greater emphasis on heredity than is
generally conceded.

In the field of phonetics, Jones (7) has shown the usefulness of
the phonetic kymograph in analyzing speech sounds as to the pres-
ence or absence of voice and nasality, in measuring the length of
sounds, and in determining the pitch of the voice. Scripture (16)
has reported a case of speech without use of the larynx. A girl
who had undergone tracheotomy at the age of three and was unable
to make any air pass through the larynx could nevertheless speak
distinctly in a faint almost toneless voice. Phonautograph records
showed the efforts at intake of air as a preliminary to speech, but
thereafter fairly regular waves comparable to those produced by the
vibrations of the vocal cords in normal subjects. Observation
revealed that in this case the subject produced “voiced” sounds by
compressing the air in the pharynx and then allowing it to escape
between the tongue and the velum, the surfaces of which were set
in vibration and thus functioned as an artificial glottis. Peters (14)
has investigated the influence of Sievers’ “visual signals’? and hand
gestures on speech melody. Eight observers read a strophe of poetry
four times, each time with a different combination of visual signals
and hand movements. Both phonograph and laryngeal records
were taken, the latter being then analyzed for the pitch relations.
Hand movements in a plane parallel to the body surface were
accompanied by relatively small variations in pitch; those in the
direction of vision were accompanied by greater variations in pitch.
Gestures toward the right and upward were accompanied by rising
tone; those toward the left and downward by falling tone. With
the use of the “Dur” signal smoother melodies occurred; with the
use of the ‘‘ Moll” signal, more broken melodies.
THE PSYCHOLOGY OF LANGUAGE 493

A large number of studies of the language development of chil-
dren have appeared in the last five volumes of the Pedagogical
Seminary. Magni (9) points out that actual recording of the vo-
cabularies of individuals has shown that the earlier estimates of
such vocabularies were far too low. Thus it was estimated that
persons of fair intelligence have a vocabulary of about 3,000 or
4,000 words, while actual counting has demonstrated that children
in the second grade have an average vocabulary of 4,480, and that
this increases to 12,000 in the eighth grade and 20,120 in college.
The vocabularies of children show great individual differences.
The results of investigation indicate an average vocabulary of 400
or 500 words for a child four years old, although for some children
the figure is much larger. Up to the sixth year, girls seem to sur-
pass boys in vocabulary, but thereafter the boys seem to excel the
girls. Nice (11) reports the speech development of a girl to the
sixth year. She declares that “E. went through no ‘parrot’ stage,
never repeating words after her parents. She never merely imitated
words but used them when she needed them.” The child’s speech
began at 14 months with three words; at 17 months she had a
vocabulary of 46 words; at 18 months, 133; at 3 years, 1,139;
at 4 years, 1,765; at § years, 2,502; at 6 years, 3,075. There was
a progressive decrease in the proportion of words relating to personal
experiences and indoor environment, and a progressive increase in
those having to do with outdoor environment and abstract terms.
The first sentence was heard at the end of the eighteenth month,
but until after the second year the sentences lacked connectives,
such as conjunctions, prepositions, articles, and the verb to be.
Oatman (13) reports a boy of 18 months with a vocabulary of over
100 words. His first sentence was heard early in the eighteenth
month. Brandenburg (4) has recorded the conversation of a girl
for an entire day at 40 months and at 52 months. At 40 months
the child used during one day 34 per cent. of her total vocabulary
of 2,500 words, speaking a total of 11,623 words in 1,873 sentences,
an average of 6.6 words per sentence. At 52 months she used
during one day 24 per cent. of her total vocabulary of 4,200 words,
speaking a total of 14,930 words in 1,967 sentences, an average of
7.5 words per sentence. Of the different words used, nouns were
most numerous (42.3 per cent.) and verbs next (30.7 per cent.).
Nice (12) recorded the conversation of a four-year-old girl for one
day. The child used 54.5 per cent. of her total vocabulary of 1,135
words, speaking a total of 10,511 words in 2,686 sentences, an
494 ERWIN A. ESPER

average of 3.9 words per sentence. Of the different words used,
the proportions of the various grammatical categories were about
the same as those found by Brandenburg, but for the total words
used the proportions were different, 28.2 per cent. being verbs, 22.1
per cent. pronouns, and 19.5 per cent. nouns. Bateman (3) gives
statistics from his own investigations and those of others as to the
time of advent of the first spoken word of the child. The data
pertain to 35 children, 18 of whom are English-speaking, 12 German,
and 5 French, Belgian, or Polish. The ages of the children at the
time of the first word vary from 8 to15 months. Of the 35 children,
I5 or 42.85 per cent. spoke their first word in the period from 10
to 11 months, while 26 or 74.28 per cent. had begun articulate
speech by the end of the first year. The boys in general seemed
to begin later than the girls. The first words are with few excep-
tions nouns or interjections.

The languages of primitive peoples have received considerable
attention. Magni (9) points out that their vocabularies have
proved much larger than was formerly supposed. Thus the vocab-
ularies of various American Indian languages have been found to
contain from 10,000 to 40,000 words. Dornan (5) has investigated
the language of the Bushmen of the Tati district of South Africa.
Their language, called Sesarwa, contains three tones, by which
words of similar sound but different meanings are distinguished.
The list of sounds includes four “‘clicks.” The author collected a
vocabulary of 2,000 words, but does not believe that this is complete.
Morphologic relations are shown mostly by suffixes and infixes.
The noun has two numbers, but no genders or cases. Abstract
words are rare. The numerals are from one to five. Verb forms
are well developed; there are two voices, five moods, and a large
number of tenses, with special forms for the negatives of these
tenses. There is very little syntax. Kroeber (8) gives an account
of the speech of a Zuni boy in his 23d and 24th months. The
first week of observation revealed a vocabulary of about 18 words,
six of which were nouns denoting consanguinity, and one a verb.
At this time the child had not yet learned to articulate a number of
Zuni sounds, such as the palatal stops, the affricatives ts and tc,
the fricatives s and c, surd L, and a series of glottalized sounds.
“So far as they occur, these are precisely the sounds with which
English-speaking children have difficulty.” Near the beginning of
the child’s 24th month, the observer heard the first “sentence,”
a noun and verb used together. During the six weeks preceding
THE PSYCHOLOGY OF LANGUAGE 495

his second birthday, the child increased his vocabulary by half or
more and learned to articulate a forward palatal stop and an
alveolar affricative, but did not acquire the fricatives s or c, surd L,
nor any glottalized consonant. “He had, however, mastered a
characteristic tonal inflection, was beginning to unvoice according
to rule, and was drifting away from his early leaning to make every
word consist of two identical but separate syllables. Of grammat-
ical structure there was as yet no trace, and sentence building
remained sporadic and of the most elementary kind.” Michelson
(10) discusses the method of approach to the problems presented
by the multiplicity of linguistic stocks on the American continent.
Chief among these problems is the question of whether this multi-
plicity is original, and the problem of the reconstruction of the
parent languages of the existing stocks. The tracing of genetic
relations between the various stocks is rendered difficult by the
lack of written records by which the historical development of the
members of the stocks might be determined. Since certain prin-
ciples of development and differentiation of languages have been
derived from the study of other stocks for which we have written
records covering long periods, ¢.g., the Indo-European stock, it
might seem permissible to apply these principles to the problems
presented by the American Indian stocks. Such a procedure
Michelson however regards as erroneous, on the ground that the
number of stocks for which we have adequate historical records is
too small to give the principles derived from their study universal
validity. Another difficulty arises from the fact that it is often
impossible to formulate rigorous phonetic laws for the American
Indian stocks, because, lacking historical records, the investigator
cannot show conclusively what disturbing factors, such as analogy,
have been at work. The resemblances between certain stocks raises
the question of whether they are due to a remote common origin
of the stocks or to borrowings. ‘There is evidence to show that
sounds, vocabulary, and syntax may be borrowed. Morphological
borrowing has some evidence in its favor, but is still a problem
awaiting solution.

REFERENCES
1. Barat. Le Langage. Rev. Philos., 1917, 83, 105-131. ;
2. Barnits, P. Les Eléments héréditaires dans le Langage. C. 7. Soc. de biol.,
1919, 82, 828-829. :
3. Bateman, W. G. Papers on Language Development. I. The First Word.
Ped. Sem., 1917, 24, 391-398.
496 ERWIN A. ESPER

OO ON

-

II.

12.
13.

14.
15.

16.

17.

. Branpenpure, G. C. & J. Language Development During the Fourth Year.

Ped. Sem., 1919, 26,'27-40.

. Dornan, $. §. The Tati Bushmen (Masarwas) and their Language. J. of Roy.

Anthrop. Instit., 1917, 74, 37-112.

. Esper, E. A. A Contribution to the Experimental Study of Analogy. Psychol,

Rev., 1918, 25, 468-487.

. Jones, D. Analysis of the Mechanism of Speech. Nature, 1917, 99, 285-287.

. Krogsper, A. L. The Speech of a Zuni Child. Amer. Anthrop., 1916, 18, 529-534.
. Maent, J. A. Vocabularies. Ped. Sem., 1919, 26, 209-233.

. Micuetson, T. Remarks on American Indian Languages. J. of Wash. Acad. of

Sct., 1917, 7, 222-234.

Nice, M. M. The Speech Development of a Child from Eighteen Months to
Six Years. Ped. Sem., 1917, 24, 204-243.

Nice, M. M. Concerning All Day Conversations. Ped. Sem., 1920, 27, 166-177.

Oatman, M. E. A Boy’s Development at Eighteen Months. Ped. Sem., 1921,
28, 52-59.

Peters, W. E. Stimmgebungsstudien. Psychol. Studien, 1918, 10, 387-572.

ScHIEFFERDECKER, P. Ueber die Differenzierung der tierischen Kaumuskeln zu
menschlichen Sprachmuskeln. Biol. Centbl., 1919, 39, 421-432.

Scripture, E. W. Speech without using the Larynx. J. of Physiol., 1915-16,
50, 397-403.

SecueHayreE, A. Les Problémes de la Langue. Rev. Philos., 1917, 84, 1-30.
SPECIAL REVIEWS

H. J. Warr. The Foundations of Music. Cambridge: University

Press, 1919. Pp. xvi + 239.

Watt’s volume on The Psychology of Sound, published in 1917,!
has been followed by the book now under consideration. It was
obvious that the new theory of consonance and interval which the
earlier study brought to light should be tried out in its application
to the art of music. In the present volume the application is
made with a confidence that leads the author to say in his Preface:
“If my results and analysis are valid, the musician should now
have a nearly complete and sure basis to work upon, that will give
a scientific foundation to all his elementary observations and satisfy
him with a sense of firm ground upon which to build” (p. ix).

In order to make the foundations of music intelligible to the
musician, the author restates his theory of sound in a clear and
well-written summary constituting the first eight chapters. This
portion of the book is commended to psychologists who may have
been somewhat appalled by the closely written and highly argu-
mentative tone of the earlier work.

The basic conception of tone is that of a “mass or volume of
minute [hypothetical] particles of sound sensation, of which those
at its centre are the most intense, while the others grade themselves
on either side in the whole volume so that the mass appears regular
and balanced” (p. 12). In addition, ‘‘we have every right to
claim that the real basis of our sense of interval is our observation
of a constant proportion between the volumes of tones” (p. 41).
This brings one to an explanation of fusional synergy in terms of
the balance established between two or more tones whose volumes
overlap, since it is probable that the higher reach of volume is the
same for all tones. While the author admits that this last inference
is not inevitable, he does not appear to consider the possibility of
any two tones of which the volumes do not overlap. Either the
upper reach of the volumic spread is identical for all at the point
of highest audibility of pitch, or else “it is conceivable that the
pyramid of tones should be acute- or obtuse-angled rather than
right-angled. But these alternatives are far from likely for various
reasons of which the most important will be set forth immediately”
(p. 10).

1 Reviewed in this journal, 1917, 14, pp. 254 ff. :

49
498 SPECIAL REVIEWS

The chief reason appears to be a second assumption regarding
the balance of fusing intervals, especially the octave. If pitch is
central to volume, the overlapping of volumes will yield a special
case when the higher volume lies just midway between the pitch
of the lower volume and their common higher terminal, giving
only an extra intensity to the upper half, and in the point of pre-
dominance that is the pitch of the higher tone. Thus “the whole
sound would approximate more nearly to the nature of a single
pure tone than would any other form of coincidence of volumes
and pitches. These things justify completely the identification of
the octave fusion with this special case of overlapping” (p. 19).

It is important to note these assumptions because the validity
of Watt’s attempt to clear away the mysteries of musical theory
is conditional upon their acceptance. Unfortunately, the one bit
of experimental evidence we have on the subject seems to run
counter to the notion of balance in the octave. The determination
of the limen for volume differences by Rich indicates that in the
middle region of the scale a just noticable difference is detected
with a constant increment of about .03 of the vibrational frequency.
Watt, to be sure, has replied to Rich’s contention that such a
progression is incompatible with the assumption that the upper of
any two tones in octave relation possesses half the volume of the
lower; but the interested reader may be left to determine the
merits of this controversy for himself.'

In applying his theory of fusion and interval to musical usage,
Watt is forced to emphasize harmonic structure at the expense of
simple melody. Thus the major portion of the book is concerned
with such troublesome restrictions of musical theory as those that
concern ‘consecutive fifths” and “a fourth from the bass.” The
conception which emerges from his analysis embraces three types
of chords and intervals. These are named, respectively, the
“symphonic,” the “paraphonic,” and the “‘diaphonic.” The first
includes the octave, the fifth, and to some extent the fourth; the
second comprises thirds and sixths; and the third, all dissonant
intervals. What may perhaps be regarded the chief contribution
of the present work is the notion of thirds and sixths as “‘fusionally
neutral.” Because of their balance the symphonic intervals tend
so strongly towards unity of impression that, when repeated, they
seem to destroy the onward movement of the melody. On the
contrary, thirds and sixths, being “neutral grades of fusion alone
do not in succession break this even flow of analytic concentration

1Cf. Jr. of Exper. Psychol., 1920, 3, 151 ff.
SPECIAL REVIEWS 499

necessary for the appreciation of the greater works in music” (p.
112). If it be argued that fifths and fourths are tolerable in primi-
tive music, Watt replies that primitive music is homophonic in
character, and is, therefore, concerned only with the “horizontal”
view of musical structure. Yet without harmony, which is the
“perpendicular” complement to the horizontal functions of melody,
we have a music which “hardly deserves the name of art. It is
merely primitive play, as it were” (p. 161). This extreme con-
clusion, which appears to deny any system to the musical usage of
the non-harmonic scales employed in the Orient, is indicative of
the influence a theory may exert over the facts it attempts to
organize.

A fourth from the bass is found to be objectionable for a different
reason—not because the fourth is too symphonic, but because its
tendency is towards paraphony or even diaphony. The fourth
Owes its position as a true consonance to its “‘cousinship” with the
fifth, but under the conditions in which a fourth from the bass is
objectionable it appears as a sort of augmented third, or paraphony,
though it may have the character of a diaphony of lesser degree of
consonance than the tritone.

But even if we accept the notion of the octave and fifth as bal-
anced and therefore symphonic units of a sort, there appears to be
no logical ground for setting off the paraphonic intervals from the
diaphonic. Our sense of proportion upon which these distinctions
are said to be based, must be remarkable indeed if we can even note
the proportional emplacements of the respective volumes and pitch
salients of two tones in the relation of a fourth. The unique neu-
trality of thirds and sixths, as against sevenths and seconds, is still
more difficult to comprehend.

Although many counter theories advanced by musicians are
carefully examined by Watt, one finds no reference to the suggestive
psychological analyses of Lipps and Meyer, or to Bingham’s clear
formulation of many of their conclusions. According to Watt the
tonic apparently derives from the fact that the attention falls most
easily upon the lowest tone, which thus becomes a point of reference
for the tones and chords of a musical unit. We might suppose the
lower tone of a sixth or a fourth to be as good a tonic as the lower
tone of a fifth, yet a simple experiment readily convinces one that
although the trend of the fifth is towards the lower of its tones, that
of the true fourth is towards the upper. If tonality is determined
by the notes corresponding to 2 or a power of 2 in the ratio numbers
of the intervals employed, then the lowest tone is not always and
500 SPECIAL REVIEWS

inevitably the tonic, and a different explanation is suggested for
the ban which music has placed upon the employment of consecutive
fifths and a fourth from the bass. The ambiguity of key or tonality
which Watt rejects as an unsatisfactory explanation of the bad-
effect produced by fifths gains significance when we know why it
is that a new tonality is announced whenever one fifth follows
another. If such consecutives are less objectionable when the
intervals lie between different voices than when they are between
the same voices, this but emphasizes the fact that the departure
from a definite tonality is more marked when the same voice is
responsible for the break. Similarly the objectionable character
of a fourth from the bass is obvious enough if one admits the tonic
trend away from the bass note, the tonality of which has hitherto
dominated, to the fourth above it which announces a new tonic;
for again we have a case of ambiguity and confusion in the musical
structure.

To be sure, this conception of the tonic raises the further question
of its origin, and implies a mode of solution which Watt rejects—
namely, the derivation of harmony and melodic trends from the
orderly arrangement of the partial tones. This is not the place to
reopen so old and debatable a question, but one may remark that
a fact so well known as the tonic trend in melody deserves more
consideration than Watt has given it.

The general attitude of the writer in formuiating scientific prin-
ciples for musical art is commendable. His last two chapters, on
the “Objectivity of Beauty” and “Aesthetics as a Pure Science,”
are convincing. In the opinion of the reviewer, Watt’s emphasis
upon volume in determining our sense of interval and musical form
gives us a newconception of sound that must be increasingly dealt
with in any attempt to trace the foundations of music. But in the
present work the observational data concerning balance among
simultaneous and successive tones are too meagre to furnish an
adequate foundation for the elaborate superstructure of musical
theory which he has reared upon them. As in the elaboration of
most theories, the “all or none” principle is unfortunately upper-
most; many things like the pattern of pure melody seem to be
neglected because they do not readily fall within the conception
dominating the author’s mind. Thus the assurance with which the
facts of musical usage are made to conform to principles which the
author has assumed makes one eager to test his assumptions.

R. M. Ocpen

CorneLt University
PSYCHOLOGICAL REVIEW FUBLICATIONS

Original contributions and discussions intended for the Psychological Review
should be addressed to
Professor Howard C. Warren, Editor PsycHotocicaL REVIEW,
Princeton, New Jersey.

Original contributions and discussions intended for the Journal of Experimental
Psychology should be addressed to
Dr. John B. Watson, Editor JouRNAL oF EXPERIMENTAL
PsycHoLoey,
244 Madison Avenue, New York.

Contributions intended for the Psychological Monographs should be addressed to
Dr. James R. Angell, Editor PsycHotocicaL MONOGRAPHS, ~
Yale University, New Haven, Conn.

Reviews of books and articles intended for the Psychological Bulletin, announce-
ments and notes of current interest, and books offered for review should be
sent to

Dr. S. I, Franz, Editor PsycHoLocicaL BULLETIN,
St. Elizabeth’s, Washington, D. C.

Titles and reprints intended for the Psychological Index should be sent to
Professor Madison Bentley, Editor PsycHoLocicaL INDEX,
University of Illinois, Urbana, Ill.

All business communications should be addressed to
Psychological Review Company
Princeton, New Jersey.
(Pp. B.)

 

PHILOSOPHICAL MONOGRAPHS
VOL. I

1. Aisthetic Experience: Its Nature and Function in Epistemology. W. D.
Furry. Pp. xvi-+ 160. $1.60. 2. The Philosophy of John Norris of Bemerton,
Frora IsapeL MacKinnon. Pp. iii-+ 103. $1.00. 3. Freedom and Purpose.
An Interpretation of the Psychology of Spinoza. James H. DunHam. Pp.
iv-+126. $1.25.

LIBRARY OF GENETIC SCIENCE AND PHILOSOPHY

Vol. 1. The Moral Life. A Study in Genetic Ethics. By Arraur ERNEST
Davies, Ph.D., Professor of Philosophy in Colorado College. xiv + 188
pages. 8vo. Cloth. Price, $1.50 net. Postage, 10 cents.

Vol. 2. Darwin and the Humanities. By James Marx Batpwin. viii 118
pages. 8vo. Cloth. Price, $1.50 net. Postage, 8 cents.

Published by

PSYCHOLOGICAL REVIEW COMPANY

Vv

(P, B.)
THE UNIVERSITY OF CHICAGO

DEPARTMENTS OF PHILOSOPHY, PSYCHOLOGY, AND EDUCATION

Courses in philosophy cover the logical, metaphysical, historical, ethical,
a and political fields, and also those of the psychology and philosophy of
religion.

Courses in psychology cover the fields of systematic, comparative, genetic,
and applied psychology. Laboratory courses giving practical training in all
phases of experimental psychology and in the construction and use of tests are
provided with adequate equipment of space and apparatus.

_, Courses in education cover the fields of history, administration and super-
vision, experimental education and measurements, and general and special
methods.

CORPS. OF INSTRUCTORS

Grorce H. Meap James H. Turts G. T. BUSWELL
Epwarp S. AMES Appison W. Moore I. N. Epwarps
Harvey CARR M. W. JERNEGAN H. C. Morrison
Forrest A. KINGSBURY Frank N. FREEMAN WALTER SARGENT
Epwarp S. RoBInson W. S. Gray S. C. PARKER
WiLiiaM BERRY Ws. REAVES J. F. Bossitt
H. R. MaysBEerry CuarLes H. Jupp F. S. BREED

A. W. KornHausER NATHANIEL BUTLER E. T. FIrBey

Special attention is invited to the work of the summer quarter, which begins
June 19th and ends September 1st. For information regarding this and other
work of the department, and also concerning fellowships, address

FOR PSYCHOLOGY AND EDUCATION, PROFESSOR CHARLES H. JUDD
FOR PHILOSOPHY, PROFESSOR JAMES H. TUFTS

 

Princeton University

Department of Psychology
Graduate Courses, 1921—1922

Experimental Psychology. Introductory course. Both terms (3). Professor
McComas and Dr. Zigler.

Research in Experimental Psychology. Both terms (3). Professor Brigham.

Psychological Seminary. Topic determined in consultation with class. Both
terms. Professors Warren and McComas.

Physiological Psychology. First term (3). Professor Brigham.
Psychopathology. Second term (3). Dr. Cotton.

The History and Psychology of Education. Second term (3). Professor
Brigham. P. B.

vi
PSYCHOLOGICAL MONOGRAPHS

VOL, XXII

93, Voluntary Isolation of Control in a Natural Muscle Group. J.C. BARNES. Pp. 50.
50 cents. 94. Psycho-Motor Norms for Practical Diagnosis. J.D. W. WALLIN. Pp. 101.
$1.00. 95. Apparatus and Experiments on Sound Intensity. A. P. Weiss. Pp. 59. 75
cents. 96. Wellesley College Studies in Psychology No. 2. Edited by Extmanor A. McC.
GAMBLH. Pp. 191. $1.75. 97%. Children’s Association Frequency Tests. H®RBDRT WooD-
Row and FrAncns LOWBLL. Pp. 110. $1.25.

VOL. XXIII

98. Scientific Study of the College Student. Harry Dexter Kitson. Pp. 81. 75
cents. 99. Whole vs. Part Methods in Motor Learning. A Comparative Study. Lovuisp
AUGUSTUS PECHSTEIN. Pp. 88. 75 cents. 100. Yale Psychological Studies, New Series,
Vol. II, No. 2, Edited by RoSweLu P: ANGIER. Pp. 159-331. $1.75. 101. The Vertical-
Horizontal Illusion. SARAH MARGARET RITTER. Pp. 114. $1.25.

VOL, XXIV
102. Two Studies in Mental Tests. I. Variable Factors in the Binet Tests. II. The
Diagnostic Value of Some Mental Tests. Cari C. BRIDGHAM. Pp. 254. $2.50. 103.
Radiometric Apparatus for Use in Psychological and Physiological Optics. C. BE. Furrep
and GERTRUDP RAND. Pp. xvi+65. 75 cents. 104, Transfer of Training and Retroaction.
Lovin WINFINLD WHBB. Pp. 90. $1.00. 105. Studies from the Psychological Laboratory
of the University of Chicago—Reliability of Mental Tests in the division of'an Academic
Group. BEARDSLEY RUML. Pp. 63. 75 cents. 106. Analysis of Mental Functions. Curr
Rospnow. Pp. 43. 50 cents. : .
VOL. XXV

107. Retroactive Inhibition as Affected by Conditions of Learning. Epwarp CHASE
ToLMAN. Pp. 50. 75 cents. 108. University of Iowa Studies in Psychology, No. 7%. CARL
BR, SnasHorp. Pp. 163. $1.75. 109. A Higher Scale of Mental Measurement and its
Application to Cases of Insanity. A. J. ROSANOFF, HELHN E. MARTIN and ISABEL R. Rosa-
NOFF. Pp. 113. $1.50. 110. An Experimental Study of Attention from the Standpoint of
Mental Efficiency. SHOAN MAsuzO FukuyA. Pp. 48. 50 cents. 111, The Interference
of Will-Impulses. ABRAHAM A. RoBacKk. Pp. 158. $1.75.

VOL. XXVI

112. Psychology of Clothing. GroRGE VAN Ness Dearborn. Pp. vi+72. $1.15. 113.
Some Imaginal Factors Influencing Verbal Expression. ESTHER H. SHaw. Pp. 187. $2.00.
114, Learning Curve Equation. L. L. TuurRsTonr. Pp. 51. 75 cents. 115. The Effect
of Alcohol on the Intelligent Behavior of the White Rat and its Progeny. ADA Hart ARLITT.
Pp. 50. 75 cents. 116, The Form of the Learning Curves for Memory. Conrap L.
KIJERSTAD. Pp. 89. $1.85. 11%. An Introspective Analysis of the Process of Comparing.
SaMUnL W. FHRNBERGER. Pp. 161. 5

 

VOL. XXVII

118-122. Psychological Studies from the Catholic University of America, Edited by
Epwarp A. Pacp. $5.50 per volume. 118, A Study and Analysis of the Conditioned Reflex.
Iqnatius A. HAmMoL. Pp. 66. 75 cents. 119. Image and Meaning in Memory and Percep-
tion. THOMAS VERNER Moore. Pp. 230. $2.70. 120. The Correlation Between Memory and
Perception in the Presence of Diffuse Cortical Degeneration. THOMAS VERNER Moorn. Pp.
49, 55 cents. 121. Clinical and Psychoanalytic Studies: i, Conscious and Unconscious Fac-
tors in Symbolism. PauL HA4NLY Furrey. ii. Hypnotic Analogies. THOMAS VERNER Moors.
iii, Concomitants of Amentia. MiRIAM EB. LOUGHLIN. Pp. 95. $1.65. 122. The Develop-
ment of Meaning. AGNES R. McDonouGH. Pp. 75. $1.35.

VOL. XXVIII

123. Quantitative Aspects of the Evolution of Concepts. CuarK L. Hutt. Pp. &5.
$1.35. 124, An Experimental Analysis of a Case of Trial and Error Learnine in the Human
Subject. Gzrorcn S. Snoppy. Pp._ 78. $1.25. 125. Work with Knowledge of Results
versus Work without Knowledge of Results. Gno. F. ARPS. Pp. 41. 65 cents. 126, Indi-
vidual Differences in Finger Reactions. EstHmr L. GATEWwooD. Pp. 43. 65 cents. 127.
‘The Lag of Visual Sensation in its Relation to Wave Lengths and Intensity of Light.
Marion A. BILLS. Pp. 101. $1.50. 128. Studies from the Psychological Laboratory of the
University of Chicago—Some Factors Determining the Degree of Retroactive Inhibition.
Epwarp Stevens RoBINsoN. Pp. 57. 90 cents. 129. The Higher Mental Processes in
Learning. JoHN C. PETmRson. Pp. 121. $1.80.

oe aoe F tti HaRotp R. CROSLAND. _Pp.

. ualitative Analysis of the Process of Forgetting. Bi _ i
159, 1085 of. or The Growth of Intelligence. Epcar A. Doty, Pp. 130. ie ane
Mental and Educational Measurements of the Deaf. JmANNETID Cu ssy, PRAMER ee 0
$2.00. 188, A Socio-Psychological Study of Fifty-three Supernormal Children. ILLI 5

Root. Pp. 134. $2.75. VOL, XXX.

izati i MEANS
i dardization of a Hard Opposites Test, Mani Hacky 3.
Pp. rm ac us Beales from the Psychological Laboratory of the University of Chi-

i rning in the White Rat.
cago—The Influence of the Time Interval Upon the Rate ee rental Fane Soars

JoserH U. YarBroucH. Pp. 52. 75 cents. 136, Critical ) 2 tee
5: i i inoi Mapison Bentuey. Pp. 94. *

Psychology from the University of Illinois. Edited by pgite

The Definition of Intelligence in Relation to Modern Methods of Mental Meeriren,

LeRoy SrockTon. Pp. 116. vii
Directory of American Psychological Periodicals

American Journal of Psychology—Ithaca, N. Y.; Morrill Hall.
Subscription $6.50. 600 pages annually. Edited by E. B. Titchener.
Quarterly. General and experimental psychology. Founded 1887.

Pedagogical Seminary—Worcester, Mass.: 950 Main Street.
Subscription $5. 400 pages annually. Edited by G. Stanley Hall.
Quarterly. Pedagogy and educational psychology. Founded 1891.

Psychological Review—Princeton, N. J.: Psychological Review Company.
Subscription $4.25. 480 pages annually.
Bi-monthly. General. Founded 1894. Edited by Howard C. Warren.

Psychological Bulletin—Princeton, N. J.: Psychological Review Company.
Subscription $5. 720 pages annually. Psychological literature.
Monthly. Founded 1904. Edited by Shepherd I. Franz.

Psychological Monographs—Princeton, N. J.: Psychological Review Company:
Subscription $5.50 per vol. 500 pp. Founded 1895. Ed. by James R. Angell.
Published without fixed dates, each issue one or more researches.

Psychological Index—Princeton, N. J.: Psychological Review Company.
Subscription $1.50. 200 pp. Founded 1895. Edited by Madison Bentley.
An annual bibliography of psychological literature.

Journal of Philosophy—New York; Sub-Station 84.
Subscription $4. 728 pages per volume. Founded 1904.
Bi-weekly. Edited by F. J. E. Woodbridge and Wendell T. Bush.

Archives of Psychology—Sub-Station 84, N. Y.: Archives of Psychology.
Subscription $5. 600 pp. ann. Founded 1906. Ed. by R. S. Woodworth.
Published without fixed dates, each number a single experimental study.

Journal of Abnormal Psychology and Social Psychology—Boston.
Subscription $5. Richard G. Badger. Edited byMorton Prince.
Bi-monthly. 480 pages annually. Founded 1906. Abnormal and social.

Psychological Clinic—Philadelphia: Psychological Clinic Press.
Subscription $2.50. 288 pages. Ed. by Lightner Witmer. Founded 1907.
Without fixed dates (9 numbers). Orthogenics, psychology, hygiene.

Training School Bulletin—Vineland, N. J.: The Training School.
Subscription $1. 160 pp. ann. Ed. by E. R. Johnstone. Founded 1904.
Monthly (10 numbers). Psychology and training of defectives.

Journal of Educational Psychology—Baltimore: Warwick & York.
Subscription $4. 540 pages annually. Founded 1gIo.
Monthly (9 numbers). Managing Editor, Harold O. Rugg.
(Educational Psychology Monographs
Published separately at varying prices. Same publishers.)

Comparative Psychology Monographs—Baltimore : Williams & Wilkins Co.
Subscription $5. 500 pages per volume. Edited by W. S. Hunter.
Published without fixed dates, each number a single research.

Psychoanalytic Review—Washington, D. C.: 3617 roth St., N. W.
Subscription $6. 500 pages annually. Psychoanalysis.
Quarterly. Founded 1913. Ed. by W. A. White and S. E. Jelliffe.

Journal of Experimental Psychology—Princeton, N. J.
Psychological Review Company. 480 pages annually. Experimental.
Subscription $4.25. Founded 1916. Bi-monthly. Ed. by John B. Watson.

Journal of Applied Psychology—Worcester, Mass.: Florence Chandler.
Subscription $4. 400 pages annually. Founded 1917.
Quarterly. Edited by James P. Porter and William F. Book.

Journal of Comparative Psychology—Baltimore: Williams & Wilkins Co.
Subscription $5. _ 500 pages annually. Founded 1921.
Bi-monthly. Edited by Knight Dunlap and Robert M. Yerkes.

vr, B viii
Cornell University Library

ental work