Limits...
The Effect of Head Orientation on Perceived Gaze Direction: Revisiting Gibson and Pick (1963) and Cline (1967)

View Article: PubMed Central - PubMed

ABSTRACT

Two biases in perceived gaze direction have been observed when eye and head orientation are not aligned. An overshoot effect indicates that perceived gaze direction is shifted away from head orientation (i.e., a repulsive effect), whereas a towing effect indicates that perceived gaze direction falls in between head and eye orientation (i.e., an attraction effect). In the 60s, three influential papers have been published with respect to the effect of head orientation on perceived gaze direction (Gibson and Pick, 1963; Cline, 1967; Anstis et al., 1969). Throughout the years, the results of two of these (Gibson and Pick, 1963; Cline, 1967) have been interpreted differently by a number of authors. In this paper, we critically discuss potential sources of confusion that have led to differential interpretations of both studies. At first sight, the results of Cline (1967), despite having been a major topic of discussion, unambiguously seem to indicate a towing effect whereas Gibson and Pick’s (1963) results seem to be the most ambiguous, although they have never been questioned in the literature. To shed further light on this apparent inconsistency, we repeated the critical experiments reported in both studies. Our results indicate an overshoot effect in both studies.

No MeSH data available.


A schematic illustration of the experimental setup for the Gibson and Pick (1963) replication experiment. The looker (left) and participant (right) were facing each other at a distance of 200 cm. The targets were located on a music stand behind the participant. The projection screen was placed behind the participant on a chair standing on a table to make it clearly visible to the looker. The mirrors were located at exactly 30° to the left and right of the looker. Note that, for illustration purposes, the relations between targets and mirrors are not proportional to the actual set-up.
© Copyright Policy
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC4979522&req=5

Figure 2: A schematic illustration of the experimental setup for the Gibson and Pick (1963) replication experiment. The looker (left) and participant (right) were facing each other at a distance of 200 cm. The targets were located on a music stand behind the participant. The projection screen was placed behind the participant on a chair standing on a table to make it clearly visible to the looker. The mirrors were located at exactly 30° to the left and right of the looker. Note that, for illustration purposes, the relations between targets and mirrors are not proportional to the actual set-up.

Mentions: Our experimental set-up is depicted in Figure 2. We aimed at mirroring the set-up as reported by Gibson and Pick (1963) as closely as possible. The looker and participant were seated in a chair and facing each other. The lookers’ chair could be adjusted in height to be in line with the participants’ eyes. All targets were placed behind the participant and attached horizontally to a music stand. The music stand could be adjusted in height such that target number 4 would always be aligned with the participants’ bridge of the nose. Targets were separated 2.86° of visual angle from the lookers’ point of view (i.e., 2.86°, 5.72°, and 8.58° to the left and right of the participants’ bridge of the nose). Because it was unclear how the looker from the original study achieved reliable head rotation, we used two mirrors that were placed at exactly 30° to either direction. If the looker had to rotate his/her head on a particular trial, he/she had to look straight into the mirror which coincided with that particular head orientation. All experimental conditions were presented to the looker through a computer screen that was placed on a chair on a table above and behind the participant. On each trial, a target number was shown on the screen together with an arrow if the participant had to rotate his/her head on that particular trial. Stimulus presentation and response recording was achieved through custom software written in Python and relying on the PsychoPy software package (Peirce, 2007, 2009).


The Effect of Head Orientation on Perceived Gaze Direction: Revisiting Gibson and Pick (1963) and Cline (1967)
A schematic illustration of the experimental setup for the Gibson and Pick (1963) replication experiment. The looker (left) and participant (right) were facing each other at a distance of 200 cm. The targets were located on a music stand behind the participant. The projection screen was placed behind the participant on a chair standing on a table to make it clearly visible to the looker. The mirrors were located at exactly 30° to the left and right of the looker. Note that, for illustration purposes, the relations between targets and mirrors are not proportional to the actual set-up.
© Copyright Policy
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC4979522&req=5

Figure 2: A schematic illustration of the experimental setup for the Gibson and Pick (1963) replication experiment. The looker (left) and participant (right) were facing each other at a distance of 200 cm. The targets were located on a music stand behind the participant. The projection screen was placed behind the participant on a chair standing on a table to make it clearly visible to the looker. The mirrors were located at exactly 30° to the left and right of the looker. Note that, for illustration purposes, the relations between targets and mirrors are not proportional to the actual set-up.
Mentions: Our experimental set-up is depicted in Figure 2. We aimed at mirroring the set-up as reported by Gibson and Pick (1963) as closely as possible. The looker and participant were seated in a chair and facing each other. The lookers’ chair could be adjusted in height to be in line with the participants’ eyes. All targets were placed behind the participant and attached horizontally to a music stand. The music stand could be adjusted in height such that target number 4 would always be aligned with the participants’ bridge of the nose. Targets were separated 2.86° of visual angle from the lookers’ point of view (i.e., 2.86°, 5.72°, and 8.58° to the left and right of the participants’ bridge of the nose). Because it was unclear how the looker from the original study achieved reliable head rotation, we used two mirrors that were placed at exactly 30° to either direction. If the looker had to rotate his/her head on a particular trial, he/she had to look straight into the mirror which coincided with that particular head orientation. All experimental conditions were presented to the looker through a computer screen that was placed on a chair on a table above and behind the participant. On each trial, a target number was shown on the screen together with an arrow if the participant had to rotate his/her head on that particular trial. Stimulus presentation and response recording was achieved through custom software written in Python and relying on the PsychoPy software package (Peirce, 2007, 2009).

View Article: PubMed Central - PubMed

ABSTRACT

Two biases in perceived gaze direction have been observed when eye and head orientation are not aligned. An overshoot effect indicates that perceived gaze direction is shifted away from head orientation (i.e., a repulsive effect), whereas a towing effect indicates that perceived gaze direction falls in between head and eye orientation (i.e., an attraction effect). In the 60s, three influential papers have been published with respect to the effect of head orientation on perceived gaze direction (Gibson and Pick, 1963; Cline, 1967; Anstis et al., 1969). Throughout the years, the results of two of these (Gibson and Pick, 1963; Cline, 1967) have been interpreted differently by a number of authors. In this paper, we critically discuss potential sources of confusion that have led to differential interpretations of both studies. At first sight, the results of Cline (1967), despite having been a major topic of discussion, unambiguously seem to indicate a towing effect whereas Gibson and Pick’s (1963) results seem to be the most ambiguous, although they have never been questioned in the literature. To shed further light on this apparent inconsistency, we repeated the critical experiments reported in both studies. Our results indicate an overshoot effect in both studies.

No MeSH data available.