Sessions and Events > Program Schedule > Symposium

Symposium 1

 

Friday June 20, 10:30AM-12:30PM at Alexander Hall

Implicit Processing and Awareness
Implicit processing and awareness are usually concomitant and functioning in harmony for normal cognitive systems. Indeed, understanding implicit processing can provide great insights into the nature of mechanisms underneath awareness, and studies on the relationship between implicit processing and awareness can elucidate the architecture of cognitive processes and the nature of consciousness. This symposium brings together researchers who tackle these issues from electrophysiology, brain imaging, neuropsychology, behavioral measures, and computational modeling. Each speaker would report experimental data taken from non-human animals, normal participants, neuropsychological patients, and neural network models to put forward hypotheses about the interplay between perception, memory, attention, and/or action in uni-sensory and multi-sensory domains.

 

Chair: Su-Ling Yeh
Department of Psychology
National Taiwan University
TAIWAN
Email: suling@ntu.edu.tw

 

Explicit vs. Implicit Perception and Working Memory Processes

Shaul Hochstein
Life Sciences Institute and Neural Computation Center
ISRAEL
Email: shaul@vms.huji.ac.il
Website: http://titanus.roma1.infn.it/CYM/CYM18/shaul_hochstein/shaul.html
Shaul Hochstein, Life Sciences Institute & Neural Computation Center, ISRAEL
Volodya Yakovlev, Life Sciences Institute & Neural Computation Center, ISRAEL
Sandro Romani, Dip. di Fisiologia Umana Universita di Roma La Sapienza, Roma, ITALY
Daniel Amit, Racah Institute of Physics, Hebrew University, Jerusalem, ISRAEL;
____________Dip. di Fisica, Universita di Roma La Sapienza, Roma, ITALY

When encountering someone on the street, we can often say whether his or her face is familiar, though it may be difficult to identify the same person. Familiarity memory may differ essentially from identification memory, which includes contextual and episodic information and may require conscious scrutiny. We tested macaque monkeys on a delayed-match-to-multiple-sample task, with limited sets of well-trained images or neverbefore- seen images. They performed better with novel images, detecting familiarity vs. novelty, rather than recency of presentation. This implies extremely effective one-shot learning of familiarity, resembling Standingˇ¦s (1973) finding that people detect familiarity for 10,000 once-seen pictures. We suggest that with conscious scrutiny delayactivity working memory is used for identifying well-trained stimuli. Novel stimuli do not induce delay-activity, so a different strategy is used, based on modulated responses to repeated images. We present a generic neural network model, quantitatively simulating these behaviors, based on conservative Hebbian synaptic plasticity. Familiarity becomes the first step toward establishing identification.
[ top ]
 
Interactions between Implicit Processing and Working Memory Revealed through Visual Extinction

Glyn Humphreys
Behavioural Brain Sciences
School of Psychology
University of Birmingham
UK
Email: G.W.Humphreys@Bham.ac.uk
Website: http://psg275.bham.ac.uk/bbs/humphreysg.htm
In the phenomenon of visual extinction, patients are unaware of a stimulus presented on the side of space contralateral to their lesion when a competing stimulus occurs simultaneously on the ipsilesional side. The phenomenon is classically associated with damage to psoterior parietal cortex. The degree of extinction manifested in patients, however, vary on the basis of whether the contralesional and ipsilesional stimuli group. This is consistent with there being implicit processing of contralesional stimuli, which enables them to enter into grouping relations with ipsilesional events. Recently we have shown that extinction can also be modulated by matches between the contralesional stimulus and items held in working memory, indicating that working memory interacts with implicit processing of items to enable stimuli to enter into awareness. We have examined the neural basis of this interaction using fMRI, with the data revealing that it is dependent on a fronto-thalamic-visual circuit, by- passing posterior parietal cortex. This suggests that deficits in realising awareness can be overcome by recruiting corticalsubcortical circuits that route information to frontal cortex.
[ top ]
 
The Colavita Effect: An Example of Crossmodal Extinction in Normal Participants?
Charles Spence
Department of Experimental Psychology
University of Oxford
UK
Email: charles.spence@psy.ox.ac.uk
Website: http://psyweb.psy.ox.ac.uk/xmodal/members.htm

Colavita (1974) first reported that presenting a light at the same time as a clearly suprathreshold auditory target resulted in many people simply failing to respond to (or be aware of) the sound (a sound that participants were always aware of when presented in isolation). I will describe a number of recent studies on this little-studied, but fascinating, crossmodal phenomenon. I will highlight the spatiotemporal constraints on the Colavita effect and its sensitivity to manipulations of attention/perceptual load. I will also describe the latest research demonstrating that the presentation of a visual stimulus leads to a significant decrease in peopleˇ¦s sensitivity to a simultaneously-presented auditory or tactile stimulus. These results, which contrast with the commonly-held view that multisensory integration always leads to perceptual/performance enhancement, have led us to argue that the Colavita effect can be thought of as a form of crossmodal extinction (experienced by stroke patients) in normal participants.
[ top ]
 
Coding of Cognitive Control Demand in the Medial Prefrontal Neurons
Keiji Tanaka
RIKEN Brain Science Institute
JAPAN
Email: keiji@postman.riken.jp
Website: http://www.brain.riken.jp/labs/cbms/tanaka.html
Resources for top-down attention are limited and therefore allocated to cognitive processes according to demand. While the anterior cingulate cortex is thought to represent such demand for attention allocation, it is unclear what determines the demand represented there. Here we report that neurons in the anterior cingulate cortex showed prominent activities between action completion and visual feedback in an action-learning task. The activities had properties of top-down attention control signals. Their magnitude correlated with the expected size of prediction errors of action values along the action learning in each block. These findings suggest that in the context of action learning the anterior cingulate cortex represents the expected size of prediction errors to indicate how much attention to pay to the action outcome.
[ top ]