Induction of novel colour categories in a non-categorical colour space

 

Belinda Gargaro, Simon Cropper and Michael Johnston

University of Melbourne

belinda_gargaro@hotmail.com

 

Friday, Commerce 2.20, 9.30 am

 

The category boundary effect (CBE) is an increase in discriminability across a perceptual boundary relative to that within a category. This study quantified the CBE for pre-existing and novel categories within a non-categorical (cardinal) colour space. This required the following five stages: 1) Four observers freely categorised the equiluminant plane of cardinal space. This allowed the identification of subjective category and boundary regions. 2) Discrimination was measured within and across subjective categories. No CBEs were apparent for any observer. 3) A subjective category was selected and a novel boundary induced through training. 4) Discrimination was retested as in stage 2. Once again no CBEs were found. 5) After no less than four weeks observers repeated stage 1. Free categorisation revealed the presence of the induced novel categories suggesting categorical colour perception in the absence of any CBEs.

 

 

 

 

Monocular texture and binocular slant

 

Barbara Gillam, Tatjana Seizova-Cajic and Michael Pianta

University of New South Wales, University of Sydney, University of Melbourne

b.gillam@unsw.edu.au

 

Friday, Commerce 2.20, 4.00 pm

 

We show how the presence of monocular texture at the sides of random dot stereograms of a horizontally slanted surface provides potential information about the slant as well as the azimuth of the surface. Experiments were conducted to compare the use of this information when the monocular texture was consistent with part of the background, with an aperture in front of the surface and with a mixture of the two. Only the latter two conditions gave a significant enhancement of slant relative to the disparity gradient alone and this enhancement was found to decrease with the width of the surface and increase with the disparity gradient. The sidebands alone had little effect on slant. The reasons for these results will be discussed. It is concluded that the process underlying slant enhancement differs from Panum's limiting case and that it involves an interaction of the monocular sidebands with the disparity gradient rather than the addition of a separate cue.


Luminance and contrast as depth cues

 

Donovan G. Govan and Robert P. O'Shea

University of Otago

nemesis@psy.otago.ac.nz

 

Sunday, Commerce 2.20, 11.30 am

 

It has long been held that luminance acts as a cue for depth perception. But varying the luminance of a stimulus inevitably alters its contrast with its background. Recent research shows that contrast is a depth cue. I have distinguished two kinds of contrast, external contrast, the contrast of a stimulus with its background, and internal contrast, the contrast within the stimulus. Four observers reported the relative depth of two squares filled with horizontal sine-wave gratings. Their luminances and internal contrasts were varied along with the luminance of their background. We found internal and external contrast to be additive effects, whereby the stimulus with either a higher internal or external contrast appeared nearer. When the internal and external contrast of the squares was equated, luminance acted as an ambiguous cue, with the darker square appearing nearer for three of the four observers and farther for the other.

 

 

 

 

A limit to the perception of transparent motion based on signal-to-noise processes

 

John A. Greenwood and Mark Edwards

Australian National University

mark.edwards@anu.edu.au

 

Saturday, Commerce 2.20, 4.00 pm

 

Transparent motion occurs when two or more different motion signals are present within the same region of space. Aim: To determine the number of transparent motion signals that can be represented simultaneously by the visual system and the processes underlying this limit. Task: Observers indicated which of 2 temporal intervals contained the highest number of signal directions. Results: For short stimulus durations, observers could perceive a maximum of 2 signals simultaneously, even though all signal intensities were above unidirectional thresholds. When signal intensities in the 2-signal condition were reduced to those used in the 3-signal condition, performance declined markedly. Further, 3 signals could be perceived when the signals drove different speed-tuned global-motion mechanisms. Conclusions: Transparency limits are based upon signal-to-noise processes at the global-motion stage, with the thresholds required to detect transparent signals being higher than those required for unidirectional signals.


Monocular transparency is not a new form of unpaired stereopsis

 

Phillip M. Grove, K. R. Brooks, B. L. Anderson and B. J. Gillam

University of New South Wales

p.grove@unsw.edu.au

 

Friday, Commerce 2.20, 3.00 pm

 

Howard and Duke (2003) generated stereograms with a grey transparent square offset from a vertical bar in one eye, and a vertical bar with a gap in the other eye. They argued that these displays were "without conventional disparity" and that the metrical depth experienced was a new form of unpaired stereopsis due to "transparency rather than occlusion". Another possibility is that the perceived depth in these displays was obtained from horizontal contours. To test this possibility, we generated three displays that contained similar horizontal contour terminations, but were inconsistent with transparency. Reliable depth was seen in all stimuli. We conclude that in our stimuli, and those of Howard and Duke, transparency is not responsible for the perception of depth, which appears to be based instead on disparate horizontal contour terminations. Our results also show that disparate contours of opposite contrast polarity can generate depth.

 

 

 

 

Explicit reasoning impairs judgements of Olympic dives

 

Jamin Halberstadt and James Green

University of Otago

jhalbers@psy.otago.ac.nz

 

Saturday, Commerce 2.25, 3.00 pm

 

Previous research has established that providing reasons for one's attitudes prior to reporting them changes the attitudes for the worse, but "attitude quality" has been difficult to validate. In the current study reasoners' and nonreasoners' assessments of Olympic dives were compared to the ratings they received by the Olympic judges. Reasoners' ratings, particularly of poorer dives, corresponded worse to those of judges, even when rating them a second time under control conditions, and did not vary with individual differences in decision-making style.


The influence of illusory line motion on saccadic latency

 

Jeff Hamm1, Trevor Crawford2 and Matthew Kean1*

University of Auckland1, Lancaster University2

m.kean@auckland.ac.nz

 

Friday, Commerce 2.25, 3.00 pm

 

Peripheral flashes are known to automatically attract visual attention. If a static bar is presented with one end near the flash, illusory line motion (ILM) is perceived away from the flashed location. Prior research has shown that visual attention will track this illusory motion. Further, it is generally accepted that visual attention is directed to a location in space prior to the execution of a saccade to that location. If ILM draws attention away from the flash and towards the non-flashed location, this should slow prosaccades - which are normally rapidly initiated - and speed antisaccades - which are normally comparatively slow. (Here, prosaccades and antisaccades are defined in reference to the location of the ILM-inducing flash). Pilot results have supported the predictions. Experimental data will be presented which explore this phenomenon further.

 

 

 

 

Putting orientation in the picture: A new theory of object recognition

 

Irina M. Harris and Paul E. Dux

Macquarie University

iharris@maccs.mq.edu.au

 

Saturday, Commerce 2.03, 12.30 pm

 

Recognition costs incurred by rotated objects have often been interpreted as evidence for viewpoint-dependent recognition. We review findings which suggest that recognition is actually mediated by orientation-invariant representations and that the viewpoint-dependent effects reflect a post-recognition process of determining the object’s orientation in space. This idea is articulated in a two-stage theory of object recognition. During the first stage, the object’s identity is established by activating an object representation stored in memory. Strictly speaking, this is the recognition process, and is achieved via orientation-invariant means. However, for an object to be consciously perceived, a second stage of processing is necessary, in which the recovered identity is bound to orientation information in order to deliver a stable percept anchored in space and time.

 

 

 

 


Depth from monocular half images: Occlusion or low level processing?

 

Julie M. Harris, Laurie Wilcox and Suzanne McKee

University of Newcastle upon Tyne, York University, Smith-Kettlewell Eye Research Institute

j.harris@ncl.ac.uk

 

Friday, Commerce 2.20, 2.30 pm

 

Depth can be perceived when a single point or object is viewed in one eye, without a partner point in the other eye (Kaye, 1978). Why? We considered two models. First, monocular points in an image are consistent with occlusion situations, such as when viewing a step edge. Second, only rarely (for fixation along the midline) do points matched on the temporal (nasal) retina not have a nasal (temporal) match in the other eye. Could the depth seen be due to a specialised mechanism for seeing depth along the midline? We explored this by measuring depth identification using both stereoscopic and monocular half-images. We assessed the effect of interocular contrast and midline vs. eccentric fixation, on perceived depth. Results are not consistent with a high-level occlusion explanation, but could be explained by a coarse low-level mechanism which matches the monocular half-image to noise in the signal from the other eye.

 

 

 

 

Dissociating detection from localisation of tactile stimuli

 

Justin Harris, Thida Thein and Colin Clifford

University of Sydney

justinh@psych.usyd.edu.au

 

Friday, Commerce 2.21, 4.00 pm

 

Neurological studies (including two reports of "tactile blindsight") suggest that detecting and locating tactile stimuli are doubly dissociable processes, presumably mediated by different neural structures. We sought evidence for such a dissociation in neurologically intact participants. We compared people's accuracy at detecting versus locating a backward-masked stimulus presented to one of four fingers. When accuracy scores were converted to a bias-free measure of sensitivity (d_), subjects were found to be better at detecting than locating the stimulus. Detection was also more susceptible than localization to manipulations involving the mask: detection improved more than localization as the target-mask interval increased; and detection, but not localization, was affected by changes in mask frequency. Comparing these results with simulated data generated by computational models, we conclude that detection and localization are not mutually independent, as previous neurological studies might suggest, but rather localization is subsequent to detection in a serially-organized sensory processing hierarchy.


Pseudoneglect in the 'mental number line'

 

Markus Hausmann

Ruhr-University Bochum

Markus.Hausman@rub.de

 

Friday, Commerce 2.25, 12.00 noon

 

The left and a right hemispace are integrated by the brain to a complete representation. This spatial mapping seems not to be veridical but slightly distorted to the left of the objective midline, most likely as a result of a superiority of the right brain in spatial awareness. This can be measured with a visual line bisection task. When normal subjects bisect horizontal lines they deviate to the left of the veridical center. Line length, line position, scan direction and many other factors strongly affect the line bisection bias. Using a mental line, that is the mental number line, it is shown that the leftward distortion is not due to physical properties of the percept. The results strongly support the notion that numbers are represented mentally from left to right, and indicate that the directional bias in bisecting physical lines is representational in origin.

 

 

 

 

Configural processing in own-race and other-race face recognition

 

William G. Hayward1, Hoi-Yan Lam1, Keung-Tat Lee1, Gillian Rhodes2, Rebecca Glauert2, Emma Jaquet2 and Elinor McKone3

Chinese University of Hong Kong1, University of Western Australia2, Australian National University3

whayward@psy.cuhk.edu.hk

 

Friday, Commerce 2.03, 10.00 am

 

Many studies of expertise in face recognition have found evidence for configural processing in recognition of own-race faces. However, few studies have examined whether recognition of other-race faces shows a corresponding reduction in sensitivity for configural information. In Experiment 1, we found that Australian participants showed greater configural sensitivity for Caucasian than Chinese faces; however, Hong Kong participants showed equal configural sensitivity for both sets of faces. The Chinese face stimuli were photographs of Chinese people resident in Australia; it is not known whether the Hong Kong participants were experts with this set. Therefore, a new set of face stimuli were created using Hong Kong university students. Hong Kong and Australian participants were then tested on detection of these and Australian Caucasian faces which had undergone configural changes. The results have implications for the generality of the configural processing hypothesis.


Why we see illusory relative motion between abutting gratings

 

Trevor Hine

Griffith University

t.hine@griffith.edu.au

 

Saturday, Commerce 2.21, 9.30 am

 

A simple stimulus comprising two abutting stationary gratings can produce a strong illusion of segmentation by relative motion. A theory of this effect is developed based upon the well-known fact that the perception of motion of entire objects in the visual field arises from the integration of local motion signals. It takes into account that the accuracy of these signals is contaminated by noise: whether the noise is neural or statistical, that is the motion signal is small (for example, as in fixational eye movements) compared to the resolution of the motion detector. The model is used to explain parametric data from experiments using the simple stimulus. Results for optimal angle (at least 120 deg between local motion directions), velocity (< 4 deg/ sec) and spatial frequency (6 - 8 cpd) are predicted. Other theories (e.g., Fermüller et al. (2000) fail to predict these and other results from the simple stimulus. The theory presesented here can be extended to explain illusory relative motion in other patterns, such as the Ouchi.

 

 

 

 

Sex hormonal modulation of hemispheric asymmetries in the attentional blink

 

Antje Hollander1, Markus Hausmann1, Jeff P. Hamm2 and Michael C. Corballis2

Ruhr-University Bochum1, Universityof Auckland2

antje.hollaender@web.de

 

Saturday, Commerce 2.22, 9.00 am

 

Differences in functional cerebral asymmetries modulated by gonadal steroid hormones were examined during the menstrual cycle in women. Women were tested with a double-stream rapid serial visual presentation (RSVP) task, with one stream in each visual field, during the low steroid menses and the high steroid midluteal phase. They were required to detect a target item, and then a probe item, each of which could appear in either stream. If the probe appeared 200-500 ms after the target, its detection was impaired-a phenomenon known as the "attentional blink." This occurred in both streams in the midluteal phase, but only in the right visual field in the during menses. Thus low steroid levels appeared to restrict the attentional blink to the left hemisphere, while high levels of estradiol and progesterone in the midluteal phase appeared to reduce functional asymmetries by selectively increasing the attentional blink in the right hemisphere.


Sex differences in the salience of spatial cues used by rats in detection of a brightness change

 

Rob N. Hughes

University of Canterbury

rob.hughes@canterbury.ac.nz

 

Saturday, Commerce 2.22, 2.30 pm

 

Without training, rats can detect one of two black arms of a Y-maze that was previously white during a prior acquisition trial. Although this ability has featured as a measure of visual recognition memory in studies of retention, it seems likely that spatial cues determine rats' choices. While all earlier research has involved male rats only, there is some recent evidence of sex differences in the behaviour. It is also possible that detection of a change may be enhanced if, during acquisition, the rats are free to enter and explore both arms (thereby maximising their spatial characteristics), rather than merely inspect them through transparent barriers. Effects on responsiveness to change of the orientation of a Y maze and the nature of acquisition trials were investigated in hooded rats of both sexes. Only males were affected which suggested that they depended more on spatial cues than females to detect a changed arm.