Teaching Documents by Rachel Cooper
% % % % 17/12/2012 created by Rachel Cooper clear all clc % clear command window RandStream.setDe... more % % % % 17/12/2012 created by Rachel Cooper clear all clc % clear command window RandStream.setDefaultStream(RandStream('mt19937ar','seed',sum(100*clock))); % reset the random seed %%%%% AIMS %%%%% % Present instruction screen which asks participants to press 'z' if the % animal is cute and 'm' if the animal is not cute. % present cute and non-cute animal images one at a time. % Wait until after image has been presented before participants respond. % record which key was pressed and RT. % Add a break after 50% of trials. % Add a final 'thank you' screen. % Ask for age. % Hide cursor during trials. % Only accept keys 'z' and 'm'. % save results to a .mat file.
Papers by Rachel Cooper
Visual Cognition, 2013
Hills and Lewis (2011) have demonstrated that the own-race bias in face recognition can be reduce... more Hills and Lewis (2011) have demonstrated that the own-race bias in face recognition can be reduced or even removed by guiding participants' attention and potentially eye-movements to the most diagnostic visual features. Using the same old/new recognition paradigm as Hills and Lewis, we recorded Black and White participants' eye movements whilst viewing Black and White faces following fixation crosses that preceded the bridge of the nose (between the eyes) or the tip of the nose. White faces were more accurately recognised when following high fixation crosses (that preceded the bridge of the nose) than when following low fixation crosses. The converse was true for Black faces. These effects were independent of participant race. The fixation crosses attracted the first fixation but had less effect on other eye-tracking measures. Furthermore, the location of the first fixation was predictive of recognition accuracy. These results are consistent with an attentional allocation model of the own-race bias in face recognition and highlight the importance of the first fixation for face perception (c.f. Hsiao & Cottrell, 2008). 3 Removing the own-race bias in face recognition by attentional shift using fixation crosses to diagnostic features: An eye-tracking study The own-race bias (ORB) in face recognition is typified by more accurate and faster recognition of own-race compared to other-race faces (for a review see Meissner & Brigham, 2001). The ORB has been extensively studied, yet the mechanisms underlying it are still under debate. Many researchers propose that it is caused by expert face processing mechanisms being employed for own-race faces but not for other-race faces (e.g., Michel, Rossion, Han, Chung, & Caldara, 2006; Valentine & Endo, 1992), whereas other researchers have proposed that people are more motivated to process own-race faces more accurately than other-race faces (e.g., Sporer, 2001). While these propositions are not mutually exclusive (although a participant may have the motivation to process other-race faces accurately they also need the perceptual expertise to do so, Hugenberg, Miller, & Claypool, 2007), the precise nature of the perceptual expertise requires further elucidation. Hills and Lewis (2006) recently proposed that the reason why other-race faces are less accurately recognised than own-race faces is due to the physiognomic characteristics of faces of different races. Black African and White European faces have different physiognomic variability for different features (McClelland & Chappell, 1998): Black faces have more variability in the nose and mouth shape than the eye and hair colour, whereas White faces have the reverse pattern. Ellis, Deregowski, & Shepherd (1975) have shown that Black and White participants tend to describe faces based on these differences: White participants described the eye and hair colour more frequently than the nose and mouth shape, whereas Black participants showed the reverse pattern. Based on this evidence, Hills and Lewis (2006) used a perceptual learning paradigm to train White participants to encode the features that distinguish between Black faces (i.e., the nose and mouth) and managed to reverse the ORB and actually make their participants recognise Black faces more accurately than White faces. Hills and Lewis (2011) extended these findings by demonstrating that extensive training was not required to alter the nature of the ORB. They used fixation crosses to guide White participants' attention to either the nose or the eyes of Black and White faces. Given that the nose is more diagnostic in differentiating between Black faces, these authors hypothesised that recognition of Black faces would be superior following a fixation cross preceding the nose, whereas recognition of White faces would be inferior. Similarly, following a fixation cross preceding the eye region, recognition accuracy of White faces
have demonstrated that the own-race bias in face recognition can be reduced or even removed by gu... more have demonstrated that the own-race bias in face recognition can be reduced or even removed by guiding participants' attention and potentially eye movements to the most diagnostic visual features. Using the same old/new recognition paradigm as Hills and Lewis, we recorded Black and White participants' eye movements whilst viewing Black and White faces following fixation crosses that preceded the bridge of the nose (between the eyes) or the tip of the nose. White faces were more accurately recognized when following high fixation crosses (that preceded the bridge of the nose) than when following low fixation crosses. The converse was true for Black faces. These effects were independent of participant race. The fixation crosses attracted the first fixation but had less effect on other eye-tracking measures. Furthermore, the location of the first fixation was predictive of recognition accuracy. These results are consistent with an attentional allocation model of the own-race bias in face recognition and highlight the importance of the first fixation for face perception (cf. .
Hills, Ross, and Lewis (2011) introduced the concept that the face-inversion effect may, in part,... more Hills, Ross, and Lewis (2011) introduced the concept that the face-inversion effect may, in part, be carried by
the first feature attended to, since the first feature fixated upon is different for upright and inverted faces. An
eye-tracking study that directly assesses this hypothesis by using fixation crosses to guide attention to the
eye or mouth region of the to-be-presented upright and inverted faces was devised. Recognition was better
when the fixation cross appeared at the eye region than at the mouth region. The face-inversion effect was
smaller when the eyes were cued than when the mouth was cued or when there was no cueing. The
eye-tracking measures confirmed that the fixation crosses attracted the first fixation but did not affect
other measures of eye-movements. Furthermore, the location of the first fixation predicted recognition accuracy:
when the first fixation was to the eyes, recognition accuracy was higher than when the first fixation was
to the mouth, irrespective of facial orientation. The results suggest that the first facial feature attended to is
more predictive of recognition accuracy than the face orientation in which they are presented.
Uploads
Teaching Documents by Rachel Cooper
Papers by Rachel Cooper
the first feature attended to, since the first feature fixated upon is different for upright and inverted faces. An
eye-tracking study that directly assesses this hypothesis by using fixation crosses to guide attention to the
eye or mouth region of the to-be-presented upright and inverted faces was devised. Recognition was better
when the fixation cross appeared at the eye region than at the mouth region. The face-inversion effect was
smaller when the eyes were cued than when the mouth was cued or when there was no cueing. The
eye-tracking measures confirmed that the fixation crosses attracted the first fixation but did not affect
other measures of eye-movements. Furthermore, the location of the first fixation predicted recognition accuracy:
when the first fixation was to the eyes, recognition accuracy was higher than when the first fixation was
to the mouth, irrespective of facial orientation. The results suggest that the first facial feature attended to is
more predictive of recognition accuracy than the face orientation in which they are presented.
the first feature attended to, since the first feature fixated upon is different for upright and inverted faces. An
eye-tracking study that directly assesses this hypothesis by using fixation crosses to guide attention to the
eye or mouth region of the to-be-presented upright and inverted faces was devised. Recognition was better
when the fixation cross appeared at the eye region than at the mouth region. The face-inversion effect was
smaller when the eyes were cued than when the mouth was cued or when there was no cueing. The
eye-tracking measures confirmed that the fixation crosses attracted the first fixation but did not affect
other measures of eye-movements. Furthermore, the location of the first fixation predicted recognition accuracy:
when the first fixation was to the eyes, recognition accuracy was higher than when the first fixation was
to the mouth, irrespective of facial orientation. The results suggest that the first facial feature attended to is
more predictive of recognition accuracy than the face orientation in which they are presented.