Developing an Instrument to Identify Individual Differences in the Processing of Pictorial and Other Non-Verbal Information

Flemming Hansen, The Copenhagen School of Business Administration & Economics
Niels Erik Lundegaard, Frederiksberg Hospital
ABSTRACT - The present paper discuss recent findings regarding brain lateralization and its implications for the study of consumer behaviour. This research suggests a new view on the way in which consumers react to different media and creative solutions. It should be possible to identify individual differences in the extent to which people rely upon left or right brain informational processing.
[ to cite ]:
Flemming Hansen and Niels Erik Lundegaard (1981) ,"Developing an Instrument to Identify Individual Differences in the Processing of Pictorial and Other Non-Verbal Information", in NA - Advances in Consumer Research Volume 08, eds. Kent B. Monroe, Ann Abor, MI : Association for Consumer Research, Pages: 367-373.

Advances in Consumer Research Volume 8, 1981      Pages 367-373


Flemming Hansen, The Copenhagen School of Business Administration & Economics

Niels Erik Lundegaard, Frederiksberg Hospital

[Sponsored by the Danish Social Science Research Council.]

[Flemming Hansen is professor, ekon. dr., The Copenhagen School of Business Administration & Economics, 60 Howitz-vej, DK-2000 Copenhagen, Denmark. Niels Erik Lundsgaard is cand. merc et mag. art. (psych), Frederiksberg Hospital. Request can be made to the first author.]


The present paper discuss recent findings regarding brain lateralization and its implications for the study of consumer behaviour. This research suggests a new view on the way in which consumers react to different media and creative solutions. It should be possible to identify individual differences in the extent to which people rely upon left or right brain informational processing.


Hemispherial lateralization implies that individuals use the left and the right brain differently. This has been studied for the last two decades. The work by psychiatrists (Wexler 1980), by psychologists (Gazzaniga 1977, Kimura 1973, and Deglin 1979) and by bra/n researchers (Wittrock 1977) has improved our understanding of the way in which the right and the left brain differ in their handling of incoming information and in controlling behaviour.

Several aspects of this specialization of the brain halves are now generally agreed upon. Whereas the left hemisphere is primarily responsible for traditional cognitive activities relying upon verbal information, symbolic representation, sequential analysis, and with the ability to report consciously of what is going on; the right brain -without the individual being able to report verbally about it - is more concerned with pictorial, geometric, timeless, musical and other non-verbal information. [These conclusions do not apply to approximately 25% of all left-handed individuals who have the functioning of the left and right brain reversed. All subjects known to be left-handed are excluded from the study to be reported on the following pages.]

Even though the differences often can be difficult to identify because of the intensive interaction going on between the right and the left brain half, it is well-established that some tasks lead to left brain activity, whereas others generate right brain processing. It is to be expected that particularly pictorial, musical and other non-verbal forms of communication give rise to right brain processing.

On this background, it has been proposed that even when attention - in the sense it is normally referred to in advertising research - is not present, it is still possible for the individual to receive some information, and to store this (Krugman 1977). Moreover, it is proposed that this process is particularly efficient with pictorial material and that the information is stored in a holistic unedited, non-verbal fashion very different from the way we normally store verbal and similar information, and is not recallable, but is easily triggered by recognition.

Individual Differences

Various authors have pointed at the possibility of individual differences in the extent to which people rely upon the left or the right brain in coping with their environment. Bogen (1977) has suggested that such differences may depend upon the nature of the stimulation the child receives, while it grows up. Children growing up in a heavy verbally dominated culture, are more likely to rely more heavily upon left brain processing. Lindzay & Norman (1977) suggest that whereas intelligence as measured by existing IQ-testing procedures mostly reflect left brain activity, some aspects of creativity may imply important right brain processing. In line with this psychiatrists (Wexler 1980) are convinced that several psychiatric disorders are explainable in terms of unbalanced functioning of the brain halves. For example, schizophrenia is believed to imply reduced left brain activity.

Even consumer psychologists have touched upon the possibility of individual differences in hemispherial lateralization. Appel, Weinstein, and Weinstein (1979), in an attempt to find hemispherial specialization in handling various informational inputs from television commercials, accidentally find significant individual differences in the way in which the test material is being handled. (for a detailed discussion of their findings see Krugman 1980).

To the extent that such individual differences exist it would imply that the right brain dominated individuals would rely more on pictorial non-verbal information and that they handle information very differently from left brain dominated people.


It is the purpose of the project being reported here to establish whether such differences exist among normal consumers in their use of left and right hemisphere. Moreover, the attempt is made to learn what implications these differences have for communication planning, segmentation, etc. The project attempts:

1.  To establish the existence of individual differences in the extent to which different consumers rely upon left and/or right hemispherial processing.

2.  To develop an operational procedure for measuring such differences.

3.  To examine the implication of such differences in terms of differences in consumer's receiving, handling, storing, and reacting to different kinds of information.

The present paper concerns itself with tentative findings regarding the first two questions.


Most studies of hemispherial lateralization have used complicated measurement procedures. Measurements have been carried out in a clinical setting and often advanced instruments have been used.

Some of the earliest findings came from studies of patients with various kinds of brain damages (Sperry 1973). Other studies have used the EEG (Electroencephalogram) technique. Here the pattern of a- and b-waves in the left and in the right brain has been studied while the respondent is carrying out different tasks.

Also special tachistoscope procedures have been developed. Here the purpose has been to control to what side of the brain visual information is transmitted. Similarly, dichotic listening techniques have been used. Here different words, pieces of music, and other auditory stimulation are presented simultaneously to the left and to the right ear. By asking respondents about what they hear, it is possible to learn in what brain half information is being processed. Finally, psychological tests normally used for diagnostic purposes have been applied in an attempt to identify individual differences in lateralization.

All of these measurement techniques are, however, complicated, and almost impossible to apply in larger scale studies of consumer behaviour. Therefore, to be able to deal with individual differences in hemispherial lateralization in larger samples of consumers, it is necessary to develop a simpler measurement device.

Since the extent to which the individual relies on right vs. left brain informational processing can be expected to influence, his overall behaviour and possibly also his beliefs and values, it might be possible to have the individual himself reporting on the extent to which right or left brain functioning dominate. Of course, direct reporting cannot be expected to be valid, but the tendency of the individual may be revealed by his responses to constructively selected questions.

For example, one can speculate that a person relying more heavily upon information stored in the right brain tend to recognize faces better than a person relying more upon left brain information processing and storing. The latter on the other hand, might be expected to recall names better than faces. Replies to items reflecting the extent to which the individual find himself relying more or less upon the one or the other kind of information could be used for identifying individual differences. It is an important part of the present project to identify and validate such statements which could be used to identify individual differences.

Validation of such statements can be approached in two ways. It is possible to test the same items on two or more groups of individuals known to differ in the extent to which they rely upon left vs. right brain processing. Such a procedure is employed in a later stage of the present project, where measurements obtained with normal individuals are compared with similar measurements obtained from people who through clinical testing have been identified to be relatively right brain dominated. The present paper, however, reports findings alone based upon the second possible validation procedure.

With a sufficiently large number of test items developed, it is possible to validate these internally by comparing the response pattern among the items themselves. For this task factor analysis is a commonly used procedure. In addition to this, it is possible to validate the test items against various psychological tests supposed to reveal left and right brain specialization. In the present study, a selection of such tests are used against which the potential test items are validated.


The validating experiment reported here has been carried out with 50 male and female subjects recruited at a student housing complex in Copenhagen. Each respondent has been paid approx. 20 $ for his/hers participation. Since some of the tests employed are quite time-consuming each respondent went through a procedure lasting approx. 3 hours. Basically, the following three types of measures were used.

1.  Reactions to 10 clinical psychological tests

2.  Dichotic listening

3.  Answer to 80 selected test items presented on an self-administered questionnaire.

Clinical Psychological Tests

A total of 10 clinical psychological tests were selected from different sources.

To make the administration easy and to facilitate the registration of the response to the tests in a quantitatively operational way, both the procedure and the scoring was transformed [Details of the transformation procedure can be obtained from the authors.], so that each respondent answered all 10 tests within reasonable amount of time, and so that the scores could be registered in a form prepared for computer processing.

In the order to which the tests were administered, the standardized forms are described below:

1.  Word-Pair Learning. (Theilgaard 1979).  10 different word-pairs, 5 logical such as street/ car, see/fish, window/curtain, and 5 non-logical such as cap/hotdog, church/cotton, apple/pencil, were presented with each pair typed on a paper card. The respondents were instructed to try to learn all 10 word-pairs by running through the pile two times in a moderate to fast tempo. Hereafter the interviewer named the "first" word on each card, registered the number of right answers out of the 10 possible. This test was believed to measure left brain activity.

2.  Copy Recall (Freeman 1965).  A meaningful story of 44 words was read load to the respondent, who was asked immediately to retell the story with the use of the same words and sentences. The copy was divided into 18 units, and the number of correct answers was registered as a measure of left brain activity.

3.  Face Recognition (Theilgaard 1979).  the material for this test consists of 28 photos of 14 men and 14 young women. The young men are all dressed in a white shirt with a black tie and the women in white gowns. All were portraits. All pictures looked very much alike. All 28 portraits are shown as 4 rows each with 7 portraits. In the test 12 of the persons were chosen and first shown in individual presentation to the respondent. Each portrait was presented to the respondent in exactly 5 seconds, and immediately after the presentation of all 12 portraits, the 28 portraits were presented. The respondents were now asked within approx. 60 seconds to identify the 12 persons they had previously seen among the 28 portraits. The interviewer registered the number of correctly identified persons, as a measure of right brain capacity.

4.  Visual Gestalts (Andersen 1976).  The material for this test consists of 4 pieces of paper with 4 different geometrical figures: A circle, a rectangle, a triangle, and a half-circle. Four different cards each with four different gestalts printed in the geometrical figures are then presented to the respondent. Each presentation lasts l0 seconds. After each presentation the respondent is asked to draw as many of the gestalts as possible. The number of correctly drawn gestalts is taken as a measure of right brain capacity.

5.  Subtraction (Theilgaard 1979).  The respondent is asked to subtract 7 from 100 and name his result, and then to continue by subtracting 7 from the result and to go on down to zero. This is done with a fixed time-limit of 30 seconds. Subsequently the interviewer registers the number of correct subtractions completed. The subtraction test reflects left brain capacity.

6.  Cube Test (after Goldstein-Sheerer, Freeman 1965).  The material for this test consists of four cubes. They are all identical, and they have the following side colours:

1.  blue,  2.  red,  3.  yellow,  4.  white,   5.  diagonally cut in white and red,  6.  diagonally cut in yellow and blue.

Figure VII, VIII, and IX from the original test are presented to the respondent in 10 seconds each. Thereafter the respondent is asked to reproduce the figures as fast as possible. The total number of seconds it takes to reproduce the three figures is registered and subtracted from 180. The resulting number is supposed to reflect right brain capacity.

7-8. Word Mobilization Test I. & II. (Theilgaard 1979).  I: The respondent is asked within 60 seconds to name as many things as possible that one can see in the street. The interviewer registers how many things there are.  II: The respondent is asked to name as many animals he can think of within 60 seconds. The interviewer registers the number of animals named.

Both measures are taken as indicators of left brain activity.

9.  Incomplete Pictures (Freeman 1965).  The 21 incomplete drawings from Wechsler's Adult Intelligence Scale are presented one by one. Because of the relative high intellectual standard of the respondents they were only permitted a few seconds to look at each drawing. As a measure of right brain capacity the interviewer registers how many errors the respondent correctly identified.

10.  Face Identification (Benton 1978).  For this test map number 7 to map number 13 from the "Benton Test" are used. Each map consists of one portrait on the top and 6 at the bottom. Out of the latter 6, 3 are portraits of the same person, as the one on the top.

On each map the respondent is asked to identify those portraits at the bottom representing the same person as the one on the top. Each map is presented for a max. of 20 seconds and the interviewer registers how many correct identification the respondent makes. This is supposed to reflect right brain capacity.

Dichotic Listening

For the dichotic listening task 78 word-pairs were selected. Examples are: RUP/BOF, STA/GJO, etc.

These word-pairs were presented to respondents simultaneously with one word to the left ear and the other to the right ear. For this purpose a stereo recorder with normal head phones was used.

In line with Frumkin (1978) the dichotic listening score was computed as the number of correct items recalled among those heard with the right ear, less the number of items recalled from those heard with the left ear divided by the total number of items recalled.

The instrument was first tested with two groups. One comprising 10 business students, and the other 10 musical students. It was believed (Sperry 1973) that the musical students would show a higher right brain score than the business students. This also was found. Most of the 10 music students scored lower than any of the business students, and the average for the two groups differ significantly. In a later test/retest reliability check the correlation between the first and second measurement for a group of business students was .76.

In the subsequent application of the dichotic listening procedure some of the test items were left out to reduce the total time spent on the listening task. This was done by deleting those items from the original version which discriminated the least between the business, and the music students. An attempt was subsequently made to validate the dichotic listening tests with two other existing tests. The first of these is normally used for psychological testing at the Copenhagen State Hospital (Heshe, et. al 1978). This attempt was not successful. Neither of the three tests correlated very well with each other. Faced with this dilemma it was decided to use the original test, since this had performed well in the first test with music and business students. This version was also technically superior.

80 Self-Administered Test Items

80 items were formulated. All of these could be answered by expressing ones view on a 5 point scale, ranging from agrees completely to disagrees completely. Some of the statements were formulated departing in what is known about lateral specialization. Examples of statements are: "I enjoy listening to music", (supposed to reflect right brain activity), "I am very good at calculating", (supposed to reflect left brain activity), "I enjoy very much watching landscape painting" (right brain), "I plan carefully how to spend my time", (left brain), etc.

In addition, items were selected from existing psychological tests. These were included based upon the assumption that the mental disorders they were supposed to measure could be related to left vs. right brain dominance. For example, a number of items used in tests of neurotic fear tendencies were included. Examples are: "I like to be told what to do", "I feel uncertain when being together with people I have not met before", etc. A total of 41 items supposed to reflect right brain dominance and 39 items reflecting left brain dominance were selected. Scoring was carried out by adding the scores on the 41 right brain items separately, and on the 39 left brain items separately. Additionally, the items were treated individually in the subsequent analysis.

Finally, for each respondent age, sex, and a couple of other demographic variables were collected.


The initial data matrix can be described as a 50 respondents by 98 items large matrix. The 98 items being:

10 test scores on the psychological clinical test

1 dichotic listening score computed as described

80 test items

1 sum-score representing the 41 right brain items

1 sum-score representing the 39 left brain items, and

5 demographic variables (sex, age, etc.)

In the course of the data analysis additional variables were generated. Those variables are described subsequently.


Clinical Psychological Tests

As it will be recalled the clinical tests were selected with five of them supposed to reflect right brain dominance, and five supposed to reflect left brain dominance. To test the interrelationship among the scores of the 10 variables, a principal component factor analysis was carried out with a subsequent vari-max rotation of the two first factors. These two factors explain 45% of the total variance. (A third factor explains additionally 10%). The results of the factor analysis is shown in table 1. It appears that 4 of the 5 test items supposed to reflect left brain dominance are associated with the first factor. Similarly, on the second factor, 4 of the right brain tests load significantly high, though 1 of them is also associated with the first factor. Neither of the remaining two tests correlate with either of the two factors.



The failure of the face identification test to correlate with any of the factors is ascribable to the fact that this test turned out to be so easy that practically all of the participants scored maximum points. Thereby, the test becomes highly insensitive. With regard to the poor performance of the word-pair learning test, one possible explanation is that this test measures general intelligence, rather than anything else.

Based upon these results, it was decided in the subsequent analysis to combine the 4 tests scoring high on the first factor into a single score reflecting each individual's tendency to rely upon left brain processing. Similarly, four tests loading high on the second factor are used to compute a combined score for each respondent. This score reflect the individual's tendency to rely upon right brain processing. Since the intervals, within which the test results vary, differ between the tests, each test score was standardized and normalized before computing the combined score.

The interrelationship between the two scores is studied in figure A. It appears that the two scores - although slightly positively related (R = 0.301)- do measure two different dimensions.



Since the purpose of the project is to differentiate between relative left and relative right brain dominance regardless of the absolute scores, a ratio was computed:

(I);   Left brain score / right brain score

               Left + right brain score

It will be seen that this is the same ratio as the one being used for the dichotic listening scores.

Dichotic Listening

The correlation between the dichotic listening score and the combined left brain activity score is .167 and only approaching significance.

The correlation between the dichotic listening score and the right brain score is -.328 and significant with p < 0.01. The resulting correlation with the ratio (I) is small and insignificant.

Even though there is some relationship between the dichotic score and the right brain score, the correlations are far below what was expected. Several reasons may be suggested for the almost complete failure of the dichotic score to validate the clinical test scores.

Basically, the dichotic listening score tests a left vs. right brain dominance. The data in figure A, however, suggested a four-way grouping rather than a two-way grouping.

A second reason for the problem may be the extreme difficulties encountered in presenting stimuli simultaneously. The same problem existed with the two other dichotic listening versions against which we originally tried to validate our own instrument. In conversations after the test, several respondents reported to have been able to hear with one ear before the other.

A third problem, of which we are aware, is more complicated. Even though information sent to the left ear is transmitted to the right brain and initially processed there, in normal individuals there is some information immediately being transmitted back to the left brain half. Moreover, it is possible that some information is picked up in the left brain, while the signals are being sent from the ear to the opposing right brain hemisphere. These complex interactions make it difficult to interpret the meaning of the dichotic listening scores. This is even more so since the disturbing processes may be more or less important depending upon whether verbal material, musical material, or nonsense words are used.

These experiences with the dichotic listening have not discouraged the research group completely, however. Further work with the dichotic listening procedure is being done in to-operation with the Frederiksberg Hospital in Copenhagen, where work is now going on in an attempt to develop more accurate procedures for the presentation of the stimuli.

The 80 Test Items

The 2 sum-scores did only correlate poorly with the clinical tests. Presumably too many irrelevant items were included. Therefore, a different analysis was carried out.

Like it was done with the clinical test the 80 test items were factor analyzed using a principal component procedure with the two first factors vari-max rotated. These two factors accounted for 18.3% of the total variance in the data. Out of the 80 items, 34 significantly correlated with the first factor, and of those 28 came from that subset of 41 which originally were believed to test left brain activity.

Similarly, among those 22 items significantly correlated with the second factor, 16 came from those 39 items originally believed to reflect right brain activity. Thus, there is some evidence that the first factor parallel the first factor of the clinical test analysis, and that the second factor reflect the same dimension as the one being measured with the second factor in the clinical test analysis. However, the low amount of variance accounted for, and the many questions not loading with any of the two factors again suggests, considerable noise in the data. Particularly, a number of items originally selected from tests of mental insecurity came out in contrast with the original hypothesis. For this reason it was decided to carry out a second analysis of the 80 scores. For this purpose the ratio (I) was used and a Pearson-moment correlation between each of the 80 tests and the score was computed.

In table 2 those 24 statements are reproduced which have significantly correlated with the ratio (I) with a probability of .85 or more. In the table the exact probabilities are reproduced.




The findings reported so far are positive in the sense that they suggest that it may be possible to develop a test battery which will throw light on the extent to which an individual is relying more or less upon left and/or right brain information processing. There are also some suggestive findings that it may be possible to develop self-administered test items which will do the same job.

There are, however, also problems with the present results. The self-administered measurement instrument has not found its final form, and elements to go into it need further validation. Within the framework of the present project several attempts are being made to this effect.

A test is being carried out with patients from a major Copenhagen Hospital's psychological ward. For these patients other evidence suggest retarded left - relatively right brain dominated information processing. Along the same lines a test is being planned with a group matching the group of university students with respect to age, and sex, but not with regard to school training, verbal experience, etc. Both of these tests are hoped to show the ability of the revised self-administered test, as well as of the 8 clinical tests to discriminate among groups with supposed different hemispherial lateralization.

Also further work is being done with the dichotic listening instrument, and it is planned to develop a tachistoscope similar to that used by Kimura (1973). Also the possibilities for using EEG-measurements for validating the presently used tests are explored.

Finally, work is going on developing new statements based upon the experience gained with the present battery and based upon renewed screening of existing findings regarding the nature of specialized left and right brain processing.

If these activities result in an improved self-administered questionnaire, it becomes possible, with this in hand, to study a number of derived hypotheses regarding the nature of left and right brain information processing. For example, it will be possible to study the extent to which pictorial information communicates better with people with a tendency to use right brain dominated information processing. Similarly, various executions of verbal material can be tested in an attempt to identify sensitivity of the information processing procedures to variations in complexity of the material.

Also with a self-administered questionnaire available, it is possible to identify segments in society with more or less left vs. right brain dominated information processing.

Similarly, it is possible to carry out choice experiments with people known to differ in their hemispherial lateralization, whereby it becomes possible to study to what an extent different choice criteria influence choices for different individuals.


Andersen, Ruth (1976), "Verbal and visuo-Spatial Memory", two clinical tests administered to a group of normal subjects. Scandinavian Journal of Psychology, 17, 196-204.

Appel, Valentine, Weinstein, Sidney and Weinstein, Curt (1979) ,"Brain Activity and Recall of TV-Advertising", Journal of Advertising Research, 19, 7-15.

Benton, A. L., Van Allen, M. W., Hansher, K. De S. and Levin H. S. (1978), Test of Facial Recognition, manual, Benton Laboratory of Neuropsychology , Division of Behavioral Neurology, University, IOWA.

Bogen, Joseph E. (1977) "Some Educational Implications of Hemispheric Specialization", in Wittrock, M. C. (ed.): The Human Brain, 133-152.

Deglin, Vadim L. (1979) "Journey through the Brain", Unesco Curier, April.

Freeman, Frank S. (1965) Psychological Testing, Holt, Rinehart and Winston, New York.

Frumkin, Lynn R., Ripley, Herbert S. and Cox, Gary B. (1978), "Changes in Cerebral Hemispheric Lateralization with Hypnosis", Biological Psychiatry, 14, 741-750.

Gazzaniga, Michael S. (1977) "Review of the Split Brain" in Wittrock, M. C. (ed.): The Human Brain, 89-96.

Heshe, Jorgen, RĂ·der, Erik and Theilgaard, Alice (1978) "Unilateral. and Bilateral ECC" Acta Psychiatrica Scandinavica, Supplementum 275.

Kimura, Doreen (1973) "The Asymmetry of the Human Brain - Recent Progress in Perception", Readings from Scientific American, San Francisco.

Krugman, Herbert E. (1977) "Memory without Recall, Exposure without Recognition", Journal of Advertising Research, 17, 7-12.

Krugman, Herbert E. (1980) "Substained Viewing of Television", Paper presented at the Conference Board, Council on Marketing Research, New York, February.

Lindzay, Peter H. and Norman, Donald A. (1977) Human Information Processing, Academic Press, New York.

Sperry, Roger W. (1973) "Lateral Specialization of Cerebral Function in the Surgically Separated Hemispheres,'' In McGuigan, F. J. and R. A. Schoonorer (eds.): The Psychophysiology of Thinking. New York, Academic Press, 209-229.

Theilgaard, Alice (1979) "Demens - Psykologisk Set", Nordisk Psykiatrisk tidsskrift, 33.

Wexler, Bruce E. (1980) "Cerebral Laterality and Psychiatry: A Review of the Literature", The American Journal of Psychiatry, 137: 3, 279-289.

Wittrock, Merlin C. (1977) The Human Brain, Englewood Cliffs, New York: Prentice Hall.