Hess Pupil Dilation Findings: Sex or Novelty?, The

Social Behavior and Personality, 1998 by Aboyoun, Darren C, Dabbs, James M Jr

Pupil dilation research initially focused upon emotional reactions and then shifted almost entirely to information processing. The present study returns to the initial research and addresses the question of whether pupil dilation in response to nudity reflects general arousal or more specific sexual arousal. Among heterosexual subjects, pictures of opposite sex nudes should produce sexual arousal, while pictures of nudes of either sex should produce a general novelty reaction. Pupil dilation was monitored among men and women viewing pictures of clothed men, clothed women, nude men, and nude women. Pupils dilated more to nude pictures than to the clothed pictures, regardless ofthe sex of the subject or the sex of the target person being viewed.

The poet Guillaume Bartas de Salluste called the eyes "windows of the soul" because of the information about internal processes that they convey (Hess & Polt, 1960, p. 349). This topic was examined in studies of pupil dilation among animals in the 1950's (Lowenstein & Loewenfeld, 1963) and among human beings in the 1960's and 1970's (Hess, 1965; Hess & Polt, 1960, 1964; Hess, Seltzer, & Shlien, 1965). The human research began with studies of emotion and shifted toward studies of information processing (Beatty & Ahern, 1979; Beatty & Wagoner, 1978). It is now generally accepted that pupil dilation increases with cognitive processing load (Beatty, 1986; Beatty & Ahern, 1979; Beatty & Wagoner, 1978; Hess & Polt, 1964).

The current study returns to emotional responses, specifically responses associated with sexual thoughts in men and women. Hess and Polt (1960) found that the pupils of women dilated more to pictures of babies, mothers and babies, and nude men, while the pupils of men dilated more to pictures of landscapes and nude women. Bernick, Kling, and Borowitz (1968) found that pupil dilation was correlated negatively with heart rate and positively with reported erection among male subjects watching a pornographic film. Hamel (1974), in a study of pupil dilation among females viewing pictures of males and females undressing, found that pictures of males with the most clothing produced the least dilation.

In this prior research, there was an emphasis upon the pupil dilation of men viewing pictures of nude women. In the present study, we measured pupil dilation in men and women viewing pictures of clothed women, clothed men, nude men, and nude women. We expected that sexual arousal might produce pupil dilation in response to opposite-sexed nude pictures, while novelty might produce pupil dilation in response to same- or opposite-sexed pictures.

METHODS

Subjects were 57 men and 50 women recruited from introductory psychology classes, mean age 21.4 years (S.D.=5.3). Four additional subjects who identified themselves as gay, lesbian, or bisexual and 26 subjects whose pupil dilation data contained extensive errors were eliminated. The problem of losing the data from so many subjects and a possible solution to this problem will be addressed in the discussion section.

Each subject viewed on a computer screen a set of 28 pictures: 7 nude females, 7 clothed females, 7 clothed males, and 7 nude males. The pictures included African-American and Caucasian males and females. The pictures were ordered randomly in sets of 4, with each set containing a nude female, a nude male, a clothed female, and a clothed male. Responses to the first 4 pictures were used for familiarization and practice and were not included in the data analysis.

The study employed a computer-controlled pupillometry system (Bernhardt, Dabbs, & Riad, 1996). Each picture was displayed for 4 sec, preceded by 4-sec display of grey, on a 15-in. monitor located approximately 18 in. in front of the subject, while the computer collected pupil diameter scores generated by an ISCAN (Cambridge, MA) pupillometry device. The system measured pupil diameter every 0.5 sec, producing 16 measurements per picture for each subject.

RESULTS

The data consisted of 16 pupil diameter measurements for each of the 24 pictures for each subject, or 384 scores per subject. Each score was converted to a percent of the subject's mean score. Scores more than 25% above or below the subject's mean (often resulting from eye blinks), were treated as missing data and set equal to 100. Scores were analyzed using a repeated measures analysis of variance that treated Subject Sex as a between- subjects factor and Picture Sex, Picture Nudity, and Time as within-subjects factors. Results of the analysis are summarized in Table 1.

The three most significant effects in Table 1 involved nudity and time: pupil size increased more over time when viewing nude than when viewing clothed stimuli. The other significant effects involved picture sex as well as nudity, reflecting especially large measures in pupil size when viewing nude males. Figure 1 shows the pattern of more dilation for nude than clothed stimuli. Figure 1 also shows the striking effect among all subjects across all stimuli of an initial sharp reduction in pupil size immediately after the onset of the stimulus.