Pupil Dilation Measures in Consumer Research: Applications and Limitations

ABSTRACT - Potential applications of pupil dilation measures to consumer research are reviewed. Although the method is a valuable measure of mental activity in consumer tasks, pupil dilations have limitations that will preclude their widespread use.

Citation:

David C. Arch (1979) ,"Pupil Dilation Measures in Consumer Research: Applications and Limitations", in NA - Advances in Consumer Research Volume 06, eds. William L. Wilkie, Ann Abor, MI : Association for Consumer Research, Pages: 166-168.

Advances in Consumer Research Volume 6, 1979      Pages 166-168

PUPIL DILATION MEASURES IN CONSUMER RESEARCH: APPLICATIONS AND LIMITATIONS

David C. Arch (Student), University of California, Los Angeles

ABSTRACT -

Potential applications of pupil dilation measures to consumer research are reviewed. Although the method is a valuable measure of mental activity in consumer tasks, pupil dilations have limitations that will preclude their widespread use.

INTRODUCTION

Eight years have elapsed since the last published review on pupil dilation measures in consumer behavior (Blackwell, Hensel, and Sternthal, 1970). The authors concluded that pupil dilations, temporary changes in pupil size, may be an effective measure of consumer response to advertising, but were unsure what processes the dilations actually measured. More recently, several trends have been noted:

* a call by a number of researchers (Payne and Ragsdale, 1978; Russo, 1978) for joint methodologies to study consumer information processing leading to product choices. The objective of multiple techniques is to improve the accuracy of process measurement.

* increased attention to eye fixations as a process-tracing device in consumer behavior (Russo and Rosen, 1975)

* a more definitive interpretation of what pupil dilations actually measure (Goldwater, 1972).

Based on these trends, it is important to update the findings of Blackwell, et. al., to determine if pupil dilations are appropriate and practical measures for consumer behavior experimentation. While pupil dilation measures might fill a void in the capabilities of current process-tracing methods, limitation of pupil dilation measures will likely prevent their widespread use.

THE NEED FOR ADDITIONAL PROCESS-TRACING METHODS

A recent review by Russo (1978) discusses five process-tracing methods currently being used in consumer behavior: chronometric analysis, eye fixations, information boards, input-output analysis, and verbal protocols. One apparent weakness in all the current methods is the failure to measure the importance consumers place on individual information inputs (ads, information board cards, etc.) as a basis for their response or choice. In fact, most methods rely on indirect measures of importance, such as the subject's self-report, or avoid the question of importance entirely. Without a true measure of importance, researchers are at a loss to explain the impact of information, one of the areas where consumer behavior studies should have important inputs for marketers and policy-makers.

PUPIL DILATION

Eye pupils have been shown to dilate in response to autonomic activity in the nervous system, with dilations ranging from 1.5 to over 9.0 millimeters. While pupil dilations have been used in a number of psychophysical experiments (see Goldwater, 1972, for a review), applications in consumer behavior have been limited and largely unsuccessful. Edgar Hess and others (Hess, 1965; Krugman, 1965) have employed pupil dilations as a measure of consumer affect to advertising. The pupils were thought to dilate in response to pleasant stimuli and contract when viewing unpleasant stimuli. Subsequent studies (Woodmansee, 1965) have found no direct relationship between pupil dilations and affect.

King (1972) used pupil size and eye direction of models depicted in advertisements as an independent variable in ad copy effectiveness study. While King's manipulations were significant, the experiment does not provide insights on the use of subject pupil dilation as a dependent variable in consumer behavior.

Meanwhile, psychophysiology experiments were establishing pupil dilation as a measure of the subjects' mental effort expended in cognitive tasks. Hess and Polt (1964) found that peak dilations of subjects was related to the difficulty of arithmetic tasks.

Kahneman and Beatty (1966) confirmed the findings of Hess and Polt in a different application, short-term memory tasks. Subjects' pupil size was sensitive to the amount of information being processed. The researchers proposed that variations of pupil size resulted from rehearsing a series of digits or words that would later have to be recalled.

Beatty and Kahneman (1966) applied similar pupil size measurements in tasks involving "more permanent" memory. Processing stages were observed in the experimental task, with peak dilations immediately following the presentation of the experimental task, and a return to normal pupil size after the stimulus was reported by the subject. Due to the wider range of dilations observed, the researchers concluded that longer-term memory tasks required greater mental effort than activities utilizing short-term memory. The subject's range of dilations was shown to diminish following extensive repetitions of a simple task, showing that learning can reduce cognitive demands.

The above studies and others cited in Goldwater (1972, p. 345) have provided support for the notion that pupil dilations reflect mental effort or activity. If mental activity could be equated with importance of the stimulus to the individual, pupil dilations would be a valuable addition to existing process-tracing methods in consumer behavior. But there are several apparent drawbacks:

* Although mental activity has important associations with importance, the relationship is not complete. For example, an unimportant stimulus could evoke substantial mental activity if it was difficult to perceive and understand. The reverse may also be true. An important stimulus might require little mental activity if it was already well-learned by the subject through prior repetition.

* Pupil dilation measures simply reflect mental activity. In themselves, no clue is provided as to what is being processed, but rather that processing is taking place. Subject protocols would also be required if the researcher wants to measure the meaning, as well as the amount, of mental activity.

One application of pupil dilation to consumer behavior does not require the effort-importance link. Public policy is concerned with optimum product information presentation to consumers. Pupil dilation measures could indicate the relative difficulty consumers experience in understanding different information presentation formats.

DESIGN PROBLEMS IN EXPERIMENTS EMPLOYING PUPIL DILATION MEASURES

Like all other physiological measures, pupil dilations are subject to a variety of distortions and interpretation problems. Tryon (1975) discusses twenty-three sources of pupil size variation, with no indication that the listing is complete. The following considerations are of particular importance:

* The light reflex is the strongest determinant of pupil size. Significant variations in the lighting used for the experiment would render the dilation results meaningless. As a pre-measure, the subject's pupil dilation should be measured under standard light changes.

* Subject factors such as age, pupil unrest (hippus), drowsiness, alertness, arousal, and physical measures such as heart rate and blood pressure have demonstrated effects on dilation. The experimenter must consider these factors in subject selection, and monitor physical signs during the experiment to aid in the interpretation of results.

* It is suggested that dilation experiments be conducted at roughly the same time of day, using subjects with the same number of hours sleep. The experimental sessions should be as brief as possible to control for subject habituation to the task.

* The amount of incentive, the degree of motor or verbal response required, and the information load resulting from the instructions to the subject should be held constant.

* The lid closure (blinking) reflex results in a momentary pupil contraction followed by a dilation. The subject should be told when to blink to control for the distortion of results.

Woodmansee (1966) states that a delayed dilation response to a stimulus may change the baseline for measuring the pupillary effects of the next stimulus. Clearly, a separation between stimulus presentation must be maintained if the response to each stimulus is to be measured.

Simpson and Paivio (1968) have found that dilation effects are increased if motor responses or verbal protocols are required concurrent with the stimulus presentation. Unless the experimenter wishes to measure the total dilation for all activities or has a proven method of partialling out these effects, any motor or verbal activity should be separated from stimulus presentation.

In summary, effective use of pupil dilation measures in a consumer behavior experiment involving visual stimuli should include:

* A constant light source. Videotape monitors or a cathode ray tube (CRT) device would be effective for this purpose, since both permit contrast adjustments to a mid-range of illumination.

* Consistent presentation formats for visual stimuli. Typed pages or transparencies would be acceptable, if the same order and organization is used for all stimuli.

* Separation of presentation and response, so that the individual dilation effects can be measured. Intervals allowing for return of pupils to baseline levels would aid in the interpretation of results.

* Control of subject factors prior to and during the experiment. Careful selection of a reasonably homogenous subject group, and proper instructions as to desired eye fixations and blinking are essential.

A high degree of equipment sensitivity is required for proper monitoring of pupil dilations. Beatty and Kahneman (1966) observed average pupil size changes of 1.1 millimeters or less in their cognitive load studies. Pupil magnification would help to correct, but not eliminate, the problem of minute measurement.

Proper monitoring of pupil dilation also requires measurement. Several types of pupillometers, commercial devices that measure pupil size, are suitable for pupil dilation experiments. Pupillometers are available that record pupil size from 20 to 30 times per second, a measurement frequency used in current physiological experiments. One manufacturer produces a device that simultaneously measures eye fixation and pupil dilations.

A dedicated computer is required to consolidate the idea produced by the dilation experiment. Several technical computers can be used in conjunction with a pupillometer, and are capable of presenting experimental output in a variety of forms ranging from raw data on pupil size to computer-plotted graphs.

POTENTIAL APPLICATIONS OF PUPIL DILATION IN CONSUMER RESEARCH

Several applications of pupil dilation in consumer behavior experiments are proposed as examples of the potential utility of the measure:

* As a measure of mental activity associated with subject's cognitive responses to persuasive messages (Wright, 1973).

* As a processing effort measure associated with verbal protocols. Subject reports of their processing activity can be associated with pupil dilations on an "on-line" basis (Payne and Ragsdale, 1978) or in the form of a "prompted protocol," where the subject verbalizes his/ her thoughts relative to a previous task (Russo and Rosen, 1975).

* As a mental activity measure associated with successive fixations in eye movement analysis (Russo, 1978).

These applications are tentative at best, with a host of experimental design questions yet to be solved.

DISCUSSION

Although pupil dilation measures have potential value in consumer research, limitations of the method are also apparent. Measurement of pupil size is an exacting science, and few researchers can afford the equipment required. Even with the most sophisticated devices, experimental conditions would have to be exceedingly precise to avoid confounding the dilation results. Verbal stimuli are preferred to visual presentations in order to avoid light reflex effects.

Obtrusive measurement of pupil size is not considered to be a problem, since pupil dilations are veridical representations of mental activity. But pupil dilation measures are totally impractical in any setting but the laboratory, restricting their research applications to lab experiments.

Most researchers, therefore, will probably settle for "second-best" measures of importance and mental activity, or beg the question entirely. Still, the potential benefits of pupil dilation measures might warrant further feasibility studies in consumer research.

REFERENCES

Jackson Beatty and Daniel Kahneman, (1966), "Pupillary changes in two memory tasks," Psychonomic Science, 5, 371-2.

Roger D. Blackwell, James S. Hensel and Brian Sternthal, (1970), "Pupil Dilation: What Does It Measure?" Journal of Advertising Research, 10, 15-18.

Roger D. Blackwell, J. S. Hensel, M. B. Phillips and B. Sternthal (1970), Laboratory Equipment for Marketing Research, Dubuque, Iowa: Kendall Hunt Publishing Co.

James F. Engel, R. D. Blackwell and David T. Kollat (1978), Consumer Behavior, Hinsdale, Ill.: Dryden Press, 352-53.

Bram C. Goldwater (1972), "Psychological Significance of Pupillary Movements," Psychological Bulletin, 77, 340-55.

E. H. Hess, and J. M. Polt (1964), "Pupil size in relation to mental activity during simple problem solving," Science, 143, 1190-92.

E. H. Hess (1965), "Attitude and Pupil Size," Scientific American, 212, 46-54.

Daniel Kahneman and Jackson Beatty (1966), "Pupil Diameter and Load on Memory," Science, 154, 1583-85.

Albert S. King (1972), "Pupil Size, Eye Direction, and Message Appeal: Some Preliminary Findings," Journal of Marketing, 36, 55-58.

H. E. Krugman (1965), "A comparison of physical and verbal responses to television commercials," Public Opinion Quarterly, 29, 323-24.

John W. Payne, E. K. Easton Ragsdale (1978), "Verbal Protocols and Direct Observations of Supermarket Shopping Behavior: Some Findings and a Discussion of Methods," in Advances in Consumer Research, Vol. 5, ed. H. Keith Hunt, Association for Consumer Research.

Berkeley Rice (1974), "Rattlesnakes, French Fries, and Pupilometric Oversell," Psychology Today, 55-59.

Edward J. Russo (1978), "Eye Fixations Can Save the World; A Critical Evaluation and a Comparison Between Eye Fixations and other Information Processing Methodologies,'' in Advances in Consumer Research, Vol. 5, ed. H. Keith Hunt, Association for Consumer Research.

Edward J. Russo and Larry D. Rosen (1975), "An Eye Fixation Analysis of Multi-alternative Choice," Memory and Cognition, 3, 267-76.

Herb M. Simpson and Allen Paivio (1968). "Effects on pupil size and manual and verbal indicators of cognitive task fulfillment," Perception and Psychophysics, 3, 185-90.

Warren Tryon (1975), "Pupillometry: A Survey of Sources of Variation," Psychophysiology, 12, 90-93.

J. J. Woodmansee (1966), "Methodological problems in pupillographic experiments," Proceedings of the 74th Annual Convention of the American Psychological Association, 1, 133-34.

J. J. Woodmansee (1965), An evaluation of pupil response as a measure of attitude toward Negroes, Doctoral Dissertation, University of Colorado.

Peter L. Wright (1973), "The Cognitive Processes Mediating Acceptance of Advertising," Journal of Marketing Research, X, 53-62.

----------------------------------------