Emotional Responsiveness to the Rorschach is Reflected in Eye Movement Responses

Dauphin, Barry, Harold Greene, Mindee Juve, and Victoria Burnett

Statement of the problem:  During the Rorschach Inkblot Method (RIM) participants make eye movements over relatively unfamiliar stimuli to mentally construct objects or descriptors that are familiar to them (Exner, 2005).  Eye movement (EM) indices provide some insight on moment to moment processing of visual information.  The indices are valuable in part, because they are difficult to censor.  Previous research utilizing eye movements has indicated that individual Rorschach cards are processed differently (Blake, 1949; Thomas, 1963; Dauphin, Greene, Juve, & Burnett, 2010).  To date, there has been no attempt to study the relationship between EM indices and variables from the Comprehensive System (CS) of scoring the RIM.  The present study begins a process of understanding the relationship among visual information processing strategies, personality, regulation of emotions, and psychopathology based on the RIM.  We attempted to discover the relationships between the individual’s affective responsiveness as reflected by the Affective Ratio (AFR) and an analogous measure of EM processes.

According to Exner (2003) the AFR is hypothesized to give some index of a person’s affective responsiveness.  The basic principle of Exner (2003) and others is that “….when the proportion of R to Cards VIII, IX, and X is high, the person is regarded as affectively responsive and, when the proportion is low, the person is affectively guarded and/or withdrawn from affective stimulation.” (p. 321).  Two indices of EM were examined during the RIM: fixation duration (FD) and saccade amplitude (SA): the distance between fixations.  Eye movement AFRs (EM-AFRs) were created by dividing the means of each EM index on Cards VIII-X by the means of each index on Cards I-VII (e.g., AFR-FD = mean FD cards VIII through X / mean FD cards I through VII).  When visual scenes are attentionally engaging, relatively longer fixation durations are expected in general (Machmann, Smith, Englert, & Henderson, 2010) as are relatively shorter saccade amplitudes (May, Kennedy, Williams, Dunlab, & Branner, 1990).  If high AFR (affective responsiveness) corresponds to being attentionally engaged, especially by the color cards, then we would expect to see a positive correlation between AFR and AFR-FD and a negative correlation between AFR and AFR-SA.


Subjects. Forty four undergraduates (ages 18 – 49) participated in the study.  All had normal or corrected to normal visual acuity.


Method. Stimuli. The 10 Rorschach blots were scanned into bitmap files for presentation on a color display CRT.  The scanned images subtended visual angles of approximately 6° by 6°.  The research team was satisfied that the scanned images and physical Rorschach blots were metameric in appearance.


Apparatus. EM indices were recorded by an Eyelink II eye tracker.  Gaze positions were sampled at 500 Hz by the video-based system that also compensated for head movements.  Gaze positions were accurate within 0.5°- 1.0° of visual angle, as reported by the Eyelink software. 


Procedure. Participants received the standard instructions for the Free Association phase of the RIM (Exner, 2005) with minor modifications to account for viewing on the computer screen.  When ready to view a card, participants were required to fixate on a central disk.  Stimulus presentations were controlled by one research assistant (RA1), and the recording of spoken responses was performed by another research assistant (RA2).

Following practice trials with geometric shapes, each subject viewed the blots in standard sequence (Blots I through X).  Trials were terminated by (i) RA1, after the subject indicated that he/she was done with spoken responses, or (ii) after 60 seconds had elapsed.  RA2 recorded the participants' spoken responses.


Results. There were significant differences among the cards on all EM indices (all p’s < .01).  This is consistent with previous findings with scanned Rorschach images (Dauphin, Greene, Juve, & Burnett, 2010).  Of particular interest is the significant correlation between the AFR and the two AFR-EM indices (FD and SA).  There was a significant positive correlation between AFR and AFR-FD as predicted with r = .39, p < .01.  There was a significant negative correlation between AFR and AFR-SA as predicted with r = -.34, p < .03.


Conclusions. The present study is noteworthy because, for the first time, we have evidence for the relationship between indices of visual information processing and CS scoring of the RIM, reflecting affective responsiveness.  We see a parallel relationship at the level of basic visual information processing strategies by participants and their verbal responses to the RIM. Thus, the larger the AFR (the more affectively responsive the individual), the longer the FD on chromatic cards and the shorter their SAs are on chromatic cards.  One of the implications of this study appears to be that EMs may be utilized to reflect indications of emotional responsiveness of individuals while viewing the RIM.  This could open up an important area of research linking cognitive psychology to the RIM.  These results and other implications for utilizing EMs for future study of the RIM will be discussed.