Daniel Starch.

Educational psychology online

. (page 11 of 41)
Online LibraryDaniel StarchEducational psychology → online text (page 11 of 41)
Font size
QR-code for this ebook

The recent investigations on the fidelity of report have been
carried out by presenting to a group of observers a series of events
either in the form of real actions or more commonly by means of a
picture. The picture is exposed to the observers for a short period,
say thirty seconds or a minute, after which they are asked to write
a report of their observations. This is usually supplemented hy
an interrogatory report consisting of answers to questions regarding
the picture. Experimental inquiry into these matters has been
stimulated primarily from the practical importance of determining
the reliability of witnesses in court. Incidentally, the results have
an exceedingly significant bearing upon the accuracy of observation
involved in learning. The main results of these experiments have
been summarized by Whipple ('lo, pp. 304 ff.) as follows:

"The chief single result of the Aussagc psychology is that an errorless
report is not the rule, but the exception, even when the report is made by
a competent S (subject) under favorable conditions. Thus, in 240 reports.
Miss Borst found only 2% errorless narratives and 0.5% errorless deposi-
tions. The average subject, when no suggestive questions are employed,
exhibits a coefficient of accuracy of approximately 75%.

"Generally speaking, attestation does not guarantee accuracy: on the
contrary, though the numbers of errors is nearly twice as great in un-
sworn as in sworn testimony (according to Stern, 1.82 times, according to
Borst, 1.89 times as great) there still remains as high as 10% error in
sworn testimony.

"Reports of children are in every way inferior to those of adults: the
range is small, the inaccuracy large, and, since the assurance is high, the
warranted assurance and reliability of assurance are both very low.
During the ages 7 to 18 years, the range, especially the range of knowl-
edge, increases as much as 50%, but the accuracy, save in the deposition,
does not increase as rapidly (20%). This development of capacity to
report is not continuous, but is characterized by rapid modification at the
age of puberty.

"The one factor that more than others is responsible for the poor
reports of children is their excessive suggestibility, especially in the
years before puberty.

"Not all the features of the original experience are reported with the
same frequency or with the same accuracy: there is, rather, a process of
selection, both in the process of observation, and also, probably, in
memory and in the formulation of the report. In general, we may say
that persons and their acts, objects, things, and spatial relations are re-


ported with consideral^lc accuracy (85-90%), whereas secondary features,
especially quantities and colors, are reported with considerable inaccu-
racy (reports on color have an error of from 40-50%).

"All authorities agree that the use of interrogatory, whether of the
complete or incomplete form, increases the range and decreases the
accuracy of the report.

"The introduction of leading or suggestive questions very noticeably
decreases the accuracy of report for children, and, unless the conditions
of report are quite favorable, even for adults. The greater suggestibility
of children is shown by Stern's results in which the inaccuracy of boys
and girls, ages 7 to 14 years, was from 32 to 39%, as against 10% in-
accuracy for young men aged 16 to 19 years."

The reasons for inaccuracies in reports of obser\'xition must be
sought in various directions. The chief ones are (i) insufficient
attention to the material observed so that only a vague impression
is produced which may easily be disturbed and modified by other
stimuli, (2) meagerness of ideas and experiences with which the
observed material may be comiected and interpreted, (3) low
retentiveness of the impressions so that other impressions can
readily distort them, (4) faint imagery in terms of which to picture
the objects, (5) lack of conscientiousness in keeping apart the
observed and the inferred items as a result of which missing or
doubtful elements are supplied by unconscious inference, (6) the
effect of suggestion through which the mind is prone to seize upon
any slight hint and to fit it into the story.

The Range of Observation. The amount of material observed
by different persons within a given limit of time varies over an
astonishingly wide range. Individual differences in the capacity
for apprehending stimuli from the outside world are probably as
large as those in other mental abilities. One person may report
several times as much of a scene or series of events as another.
The pupil with a wide scope of apprehension and observation has
a tremendous advantage over one with a narrow scope of observa-
tion. This range probably depends fundamentally upon the span
of attention, quickness of assimilation of items, retentiveness, and
previous knowledge about the facts to be observed. The span of
attention as measured by rapid exposure methods, ranges in normal
persons from three or four objects to eight or nine. Such experi-
ments are carried out by exposing momentarily before the observer
cards with varying numbers of objects and by recording the mun-
ber of objects noticed.




Some years ago a number of interesting investigations were
made by G. Stanley Hall ('S3) and others to determine the range of
observation and knowledge of pupils entering school. The returns
showed a surprisingly narrow range of information and the great
extent to which it was determined by the environment in which
the child lived. The educational value of such inquiries has been
to emphasize the importance of a wide variety of immediate con-
tact and experience with real objects by observing, manipulating
and using them.

As a concrete illustration, the author made the following simple
experiment to indicate the range and fidelity of such observations
as would be made in a class in biology. Some plants in a jardiniere
were exhibited before a class of thirty-nine students for thirty sec-
onds witli instructions that they were to observe them as carefully as
possible and that they would be asked immediately afterwards to
record their observations. The shortest report (A) was only fifty-
seven words in length, and one of the longest and best (B) was 131
words in length. One of the most erroneous ones is given under (C).
B is more than twice as complete as A, and fully as free from error.
C is quite typical with regard to kind and frequency of misstate-
ments. The erroneous portions are italicized and the corrections
are given in parentheses. All three persons had had an elementary
course in botany.

Report A:

"There was a high upright geranium plant having no buds but broad
leaves. This was surrounded by low plants with drooping stems and
bearing many pointed, small leaves. The leaves had a pinkish center,
surrounded by a pale green band, the contour very irregular and the
general effect bushy. There were no buds on these plants either."

Report B :

" From a brown jardiniere arose one stalk of a geranium bearing nine
big green leaves. The leaves spread out in all directions and are round
in shape with large scallops. Lower was another plant with much smaller
leaves and more bushy. It had three large branches, one leaning over
the pot on each side and one across the front. The leaves of this plant
were more oblong in shape, rounding at the base and reaching a point.
The color was a pale pink in the center and back to the base, shading to
a deep red or wine (pink) color towards the tip end and the whole edge of
the leaf was green. The stem was much more delicate than the stem
of the geranium and the leaves were much more numerous."


Report C:

"The plant was in a brown bowl. There were varieties of plants. The
one had one large stalk -wiikjive (nine) branches growing out of it. The
leaves were large and hearl-shapcd (rounding). The other plant was lower
and drooped, and had more leaves. The leaves were oblong (heart-
shaped) and smaller than the leaves of the first plant. They had a center
of very light green (pink) and were outlined by darker green. The leaves
were smooth and glossy (velvety). The leaves of the tall plant were

It is obvious that the reporter of B has a great advantage over
either A or C both in the quantity and in the quality of his observa-
tions. In a given period of time B will learn much more material
and assimilate it in more correct form than either A or C. The
objection might be raised that a report formulated after the obser-
vations have been made is bound to be erroneous and that it would
be fairer to have all three persons stand before the plant and re-
cord their observations at the time. To this, however, we may
reply that the range of facts observed and recorded in a given
period of time would be just as wide and that observed items are
not really assimilated mentally until they can be adequately
thought out and expressed. The simple experiment here reported
probably represents quite fairly the sort of things that occur in
ordinary observation of material in learning.

Improvement in Observation. Granting the importance of
a wide range and a high degree of accuracy in the perception of the
material to be learned, what may we do to increase the accuracy,
scope, and fidelity of observation? Probably the only advice to
give at the present time is the rather obvious suggestion: Insist
on greater accuracy and attention in observation. This may be
accomplished by definite discovery of errors and inaccuracies in
the observations themselves. Experiments have shown that many
people, especially children, do not realize their inaccuracies and
that calling attention to the discrepancies between objects and
mental impressions of them results in a material reduction in
unreUability. Whipple ('10, pp. 309 if.) has summarized the experi-
mental results on the -possibility of improvement in observation
by repeated tests with the same groups of persons as follows:

"Simple practice in reporting even without special training or conscious
effort to improve, facilitates and betters the report, as is shown in Ta-
ble 47, from Miss Borst. It will be noted that the tendency to oath and






















91 .0









98. 4























warranted tendency to oath are both particularly improved by practice,
and that there is also an appreciable improvement in range, accuracy,
warranted assurance, and reliability of assurance, whereas assurance and
accuracy of assurance are scarcely affected. Similar practice-effects
may be discerned in their deposition. From these results, it is clear that
the several coefhcients of report may vary more or less independently."

Effect of Practice upon Coefficients of Report (Narrative)


Number of Report (Test) i

Range 39 . o

Accuracy 86.6

Assurance 96 . 6

Warranted assurance 84.0

Reliability of assurance 87.5

Accuracy of assurance 97 -o

Tendency to oath 43 ■ o

Warranted tendency to oath 40. 2

Unwarranted tendency to oath. ... 2.8

Reliability of oath 93 .0

Noxr;: The effect of practice in these tests is somewhat obscured by the fact that the
first and third tests were made after a 3-day, the others after a g-day interval.

"The capacity of children to observe and report in a detailed and
accurate manner may be improved by systematic training. This educa-
tion may be best secured by appeal to zeal, interest, enthusiasm, or desire
for improvement on the part of the child; more formal training of an
intellectual type, e. g., suggestions for systematic observation, specific
training in sense -perception, mstruction designed to augment appropriate
apperceptive-masses, etc., is much less effective.

"The inadequacy of the child's report is due, not so much to poor
memory, as to the fact that he fails to perceive many features in the
original experience, that he fails to put into words even what he does
perceive, and especially to the fact that he is absurdly imcritical (his
assurance, indeed, commonly reaches 100%)."

Tests such as these, but made with the material of school studies,
would probably be very useful in bringing about more concen-
trated attention upon, and greater reliability in, observation.
Thus a plant or a flower in a course in biology, might be exposed
for a definite period of time to a class of pupils who would then be
asked to write as accurate a description of the object as possible.
This description could then be definitely compared, point by point,
with the original object and in this manner the errors and inaccu-
racies would be discovered and noted. Difficulty in acquiring the


material in various school subjects is no doubt traceable in larger
part than we realize to the incompleteness and the unreliability
of the perception of the material or stimuH to be acquired. It
would be an experiment worth making to determine to what extent
the difficulty of a pupil in learning to spell is due to actual incom-
pleteness of the observation and perception of the letters in the

Interpretation of Stimuli. To a large extent, observation is
interpretation. The same identical sense impressions are inter-
preted very differently by different observers. This may be dem-
onstrated perhaps in extreme form in such tests as the one with
the ink blots outlined in the author's Experiments, Chapter XIII.
The first ink blot in that series signified to eight persons the follow-
ing different things: map, bear, trees, lake, cloud, child, bat, man
running. The same mental processes occur in a less variable man-
ner in all kinds of observation. Incoming stimuli are interpreted
by the association processes aroused in the mind. On this basis
the traditional doctrine of apperception has been formulated and
from it have been derived such pedagogical corollaries as, "Link
the new to the old," or "Proceed from the known to the unknown."
The theory of apperception has been very clearly expressed by
James in the following statement:

"The gist of the matter is this: Every impression that comes in from
without, be it a sentence which we hear, an object of vision, or an efflu-
vium which assails our noses, no sooner enters our consciousness than it is
drafted off in some determinate direction or other, making connections
with the other materials already there, and finally producing what we
call our reaction. The particular connections it strikes into are deter-
mined by our past experiences and the 'associations' of the present sort
of impression with them. If, for instance, you hear me call out A, B, C,
it is ten to one that you will react on the impression by inwardly or out-
wardly articulating D, E, F. The impression arouses its old associates:
they go out to meet it; it is received by them, recognized by the mind as
'the beginning of the alphabet.' It is the fate of every impression thus
to fall into a mind preoccupied with memories, ideas, and interests, and
by these it is taken in. Educated as we already are, we never get an
experience that remains for us completely nondescript: it always reminds
of something in quality, or of some context that might have surrounded
it before and which it now in sonie ways suggests. This mental escort
which the mind supplies is drawn, of course, from the mind's ready-made
stock. We conceive the impression in some definite way. We dispose
of it according to our acquired possibilities, be they few or many, in



the way of 'ideas.' This way of taking in the objects is the process of
apperception. The conceptions which meet and assimilate it are called
by Herbart the ' apperceiving mass.' The apperceived impression is
engulfed in this, and the result is a new field of consciousness, of which
one part (and often a very small part) comes from the outer world, and
another part (sometimes by far the largest) comes from the previous con-
tents of the mind." ('99, p. 157.)

The doctrine of apperception as here stated by James is simply
a statement of the psychology of perception as usually accepted.
The importance of the applications of the doctrine to teaching has
perhaps been overemphasized in the educational writings of the
recent past and in the pedagogical methodology that has been
worked out in accordance with its corollaries. The applications
of the theory as represented by the injunction, "Link the new to
the old," is no doubt sound from the psychological side and useful
from the pedagogical side as a general guiding principle. Illustra-
tions of the principle would be the teaching of a tppic in geography
by connecting it up with the known facts of geography in the
immediate environment of the child, or of teaching laws of chem-
istry by relating them to familiar facts and problems that have
arisen within the child's own experience, or of teaching the spelling
of a new word by pointing out its similarities to words already
known, or of teaching forms of a foreign language by referring
them to related forms in the language previously acquired. Such
a procedure is unquestionably valuable whenever it can be em-
ployed. However, some of the enthusiastic advocates of the
doctrine have been somewhat blinded to the limitations of it. If
we regard learning as a process of establishing connections between
elements of the learning-material, we can conceive of thiree possible
kinds of bonds to be formed: (i) between two known elements
which had previously not been connected, (2) between a known
and an unknown element, (3) between two unknown elements.
It is obvious that a great deal of learning consists in the formation
of the third type of connections. The doctrine of apperception
can apply only to the first and second type of connections and
these can very probably be formed more readily according to the
natural workings of apperception because some of the elements
had previously been acquired.

Much of the discussion in favor of the doctrine of apperception is
really based upon the greater practical value of the known and
nearer at home, and upon the urgent need of knowing something


about the immediate environment rather than about a distant
time or territory, whose history or geography may be of Httle value
to the child, than upon greater ease in the formation of bonds be-
tween a known and an unknown element. The main thing in educa-
tion is not to proceed from the known to the unknown, but rather
to acquire the unknown. If this can be done by linking the new
to the old, well and good, but the chief object is the Unking of the
new. Much of the so-called proceeding from the known to the un-
known or of the linking of the new to the old, is more or less fruit-
less, since it neither proceeds to the unknown nor links anything
new. It usually consists of a reawakening of the known and of the
old. Learning is fundamentally the acquisition of new sets of
stimuh-association-response series. When a pupil first attempts to
write, he must acquire new neural connections in securing control
of his hands. When he begins to learn the meaning of printed
symbols, he is confronted with new stimuli among which new con-
nections must be established. Much of the so-called teaching ac-
cording to the theory of apperception consists in setting up prob-
lems concerning things with which the child is already familiar
and thus in arousing in him a desire to learn something new. This
is, no doubt, good teaching, but the important part in learning is the
new element to be sought and the new associations to be built up.



Problems. The chief problems to be considered are as fol-

(i) How rapidly are new associative bonds formed?

(2) Does the rate of acquiring new connections and new materials
continue uniformly per unit of time or per repetition?

(3) Do variations in rate occur in a uniform manner?

(4) What causes will bring about variations in rate?

(5) Does the rapidity of learning occur in a similar manner in
all types of learning?

Such problems as these may profitably be raised with regard to
any sort of learning. If we consider the learning of a language we
may ask, How rapid is the progress in acquiring the meanings of
the words, knowledge and use of grammatical forms and ability to
translate? Is progress uniform or are, there times of rapid advance
occurring in alternation with periods of little or no gain? What
conditions will promote the learning of the language most ef-
fectively? If we could answer these and similar questions con-
cerning any type of learning we would be able to control its
progress far more economically than we are able to at the pres-
ent time.

The Curve of Learning. The rate and progress of learning
may be expressed in terms of the amount done per unit of time, or
in terms of time required per unit of work. The relation between
these two variables is represented by the curve of learning in
which one function, usually time, is represented along the base
Hne, and the other, usually amount accomplished, is represented
along the vertical axis. Figure 36 represents a typical curve of
learning in which progress is measured by the amount achieved per
five minutes of time. It represents the rate of forming associations
between numbers and letters in transcribing letters into numbers
as specified in the author's Experiments ('17), Chapter X.

Characteristics of Learning Curves. Most of the experimental
work on the course of learning curves has been done chiefly with
various kinds of skills such as telegraphy, typewriting, tossing




l^alls, mirror tracing, substitutions, and the like. Little has been
done on the progress of analytical t}^es of learning, on the advance
in the acquisition of facts of a science or of the history of a country,



^3 220

•I 200

f^ 180


« 140

•B 120

I 100

I 80

«^ 60









\ 1















5 10 15 20

Five Minute Periods


Fig. 36. — A curve of learning showing the progress of one person in learning
to substitute numbers for letters in the experiments outlined in Chapter X,
Experiments in Educational Psychology.

or on the rate at which a child learns to read or to write. Conse-
quently most of our generalizations up to the present time have
been based upon curves that represent the acquisition of skill.

8 12

32 30

Fig. 37. — Improvement in telegraphy,
and Harter ('97, p. 49).

16 20 24 28
Weeks of Practice

Individual E. L.

B. After Bryan

Such curves seem to have in common two general characteristics,
although it is doubtful whether they are universal in all types of
learning: (i) An initial period of rapid progress, and (2) successive
periods of no progress, or plateaus, followed by periods of rapid



progress. Theoretically, a curve of learning may have two initial
directions: (i) rapid progress followed by slower progress, or (2)
slow progress followed by more rapid progress, that is, a convex or a
concave form with all possible shapes between these extremes. The
large majority of curves of learning derived to date are of the former
sort. All the illustrations reproduced in Figures 37 to 41 have this
general shape. In the substitution test referred to in Figure 36
the author found that among twenty curves obtained from as
many individuals, thirteen were of convex form, six were practically
straight lines rising from left to right, and one was of concave
form. Hence, initial rapid gain seems to be a very common feature

Weeks of Practice
Fig. 38. — Improvement in telegraphy analyzed.
Bryan and Harter v'99, P- 350).

Individual J. S. After

in curves of skill. Other types of curv^es are shown in Figures 42

Online LibraryDaniel StarchEducational psychology → online text (page 11 of 41)