Publications from 1997-1998

Titles:

Orientation During Initial Learning and Subsequent Discrimination of Faces

Effects of Orientation on Recognition of Facial Affect

Effects of Observer Orientation on Perception of Ego- and Exocentric Spatial Locations

Learning to Identify Airports from Photograps vs. Navigational maps Presented at Various Orientations

Effects of Map Orientation During Learning on Airport Identification

Orientation During Initial Learning and Subsequent Discrimination of Faces

M. M. Cohen

ABSTRACT- Discrimination of facial features degrades with stimulus rotation (e.g., the "Margaret Thatcher" effect). Thirty-two observers learned to discriminate between two upright, or two inverted, faces. Images, erect and rotated by ±45°, ±90°, ±135°, and 180° about the line of sight, were presented on a computer screen. Initial discriminative reaction time increased with stimulus rotation only for observers who learned the upright faces. Orientation during learning is critical in identifying faces subsequently seen at different orientations.

Abstracts of the Psychonomic Society, 1997, 2:71.

Effects of Orientation on Recognition of Facial Affect

M. M. Cohen, J. B. Mealey

INTRODUCTION: The ability to discriminate facial features is often degraded when the orientation of the face and/or the observer is altered. Previous studies have shown that gross distortions of facial features can go unrecognized when the image of the face is inverted, as exemplified by the "Margaret Thatcher" effect. This study examines the time needed for erect and supine observers to distinguish between smiling and frowning faces that are presented at various orientations. The effects of orientation are of particular interest in space, where astronauts often view one another in orientations other than the upright. METHOD: Sixteen observers viewed individual facial images of six people on a computer screen; on a given trial, the image was either smiling or frowning. Each image was viewed when it was erect and when it was rotated (rolled) by ±45 deg, ±90 deg, ±135 deg, and 180 deg about the line of sight. The observers were required to respond as rapidly and accurately as possible to identify if the face presented was smiling or frowning. Measures of reaction time were obtained when the observers were both upright and supine. RESULTS: Analyses of variance revealed that mean reaction time, which increased with stimulus rotation (F=18.54, df 7/15, p<0.001), was 22% longer when the faces were inverted than when they were erect, but that the orientation of the observer had no significant effect on reaction time (F=1.07, df 1/15, p>.30). CONCLUSIONS: These data strongly suggest that the orientation of the image of a face on the observer's retina, but not its orientation with respect to gravity, is important in identifying the expression on the face.

Aviation, Space, and Environmental Medicine, 1997, 68:640.

Effects of Observer Orientation on Perception of Ego- and Exocentric Spatial Locations

June Li, Malcolm M. Cohen, Charles W. DeRoshia, and Lawrence T. Guzy

Perceived eye position and/or the perceived location of visual targets are altered when the orientation of the surrounding visual environment (Cohen, Ebenholtz & Linder, 1995, Perception & Psychophysics, 57(4:433) or that of the observer (Cohen & Guzy, 1995, Aviation, Space, and Environmental Medicine, 66:505) is changed. Fourteen subjects used bite boards as they lay on a rotary bed that was oriented head-down (-) 15 and 7.5 degrees, supine, head-up (+) 7.5, and 15 degrees. In the dark, subjects directed their gaze and set a target to the apparent zenith (exocentric location); they also gazed at a subjective "straight ahead" position with respect to their head (egocentric location). Angular deviations of target settings and changes in vertical eye position were recorded using an ISCAN infrared tracking system. Results indicated that, for exocentric locations, the eyes deviate systematically from the true zenith. The gain for compensating changes in head/body orientation was .69 and .73 for gaze direction and target settings, respectively. In contrast, "straight ahead" eye positions were not significantly affected by changes in the subject's orientation. We conclude that subjects make systematic errors when directing their gaze to an exocentric location in near-supine positions. This suggests a systematic bias in the integration of extra-ocular signals with information regarding head orientation. The bias may result from underestimating changes in the orientation of the head in space. In contrast, for egocentric locations, where head/body orientation information can potentially be discarded, gaze directions were unaffected by head orientation near supine.

Perception, 1997, 26 (suppl.):102-103.

Learning to Identify Airports from Photograps vs. Navigational maps Presented at Various Orientations

M. M. Cohen, J. B. Mealey, & K. Jordan

INTRODUCTION - We previously demonstrated: 1) initial learning of navigational maps produced discriminative reaction times (RTs) that were most rapid when the maps were seen in the same orientation as that in which they were initially learned, 2) RTs were significantly reduced with repeated presentations, and 3) information learned from the maps was not sufficient for all observers to recognize aerial photographs of the same airports. The present study examined the relative effectiveness of initial training on photographs versus maps. METHOD - Each of 40 participants was trained to discriminate between two navigational maps or two photographs of airports that were seen in only one orientation; they subsequently were tested with both maps and aerial photographs of the same airports that were seen in various orientations rolled about the line of sight. RESULTS - When initially trained on navigational maps, participants responded faster to the maps than to the photographs; when initially trained on photographs, they responded more quickly to the photographs (df = 1/39; F = 32.7, p <0.001). Discriminative RT decreased with repeated trials (df="4/156;" F="111.4," p<0.001). The rate at which RT decreased with trials was greater when participants responded to new than to previously learned stimuli (df="4/156;" F="28.6," p < 0.001). The effects of orientation varied with trials, and depended on the initially trained stimuli, i.e., maps or photos (df="28/1092;" F="1.8," p < 0.01). CONCLUSIONS - The speed with which observers can identify maps or photographs of airports depends on initial learning conditions, some of which are shown here.

Paper presented at the Aerospace Medical Association's annual meeting in Seattle, May 18, 1998.

Effects of Map Orientation During Learning on Airport Identification

J.B. Mealey, M.M. Cohen, K. Jordan

BACKGROUND: If an observer first learns to recognize an object in a specific orientation, a significant increase in processing time usually occurs when the object is subsequently seen in a different orientation; this phenomenon is called the "misorientation effect." The present study examines how quickly and how accurately human observers discriminate between airport maps that are viewed in orientations other than those in which they were initially learned. METHOD: Participants were trained to discriminate between two navigation maps that were seen in only one orientation; they subsequently were tested with maps and aerial photographs of the same airports that were presented in various orientations. RESULTS AND CONCLUSIONS: There were three principal findings: First, discriminative responses to maps of airports were most rapid when the maps were seen in the same orientation as that in which they were initially learned; second, a significant reduction in reaction time (RT) occurred with repeated presentations of the misoriented stimuli; third, information learned from navigation maps was not sufficient for all observers to recognize aerial photographs of the same airports.

Aviation, Space, and Environmental Medicine, 1998, Vol. 69, No. 2, pgs 104-110