The contribution of automatic and controlled processes to cue-competition in human learning.
Lead Research Organisation:
University College London
Department Name: Psychology
Abstract
Abstracts are not currently available in GtR for all funded research. This is normally because the abstract was not required at the time of proposal submission, but may be because it included sensitive information such as personal details.
Organisations
Publications
Beesley T
(2012)
Investigating cue competition in contextual cuing of visual search.
in Journal of experimental psychology. Learning, memory, and cognition
| Description | 1. Cue-competition in contextual cuing of visual search (6 experiments; 206 experimental sessions) These experiments tested the prediction that cue-competition relies on controlled reasoning processes and should therefore not be observed in these low-level learning tasks. The results support this hypothesis: we did not observe cue-competition effects in any of the experiments. In fact, for some designs we observed the opposite effect: learning about the "to-be-blocked" cue was in fact enhanced. This novel augmentation effect suggests that learning in visual cuing is non-competitive and therefore fundamentally different to learning observed in causal learning tasks. Additional experimental work using this task examined the role of attention by making distractor sets more salient by manipulating colour. Augmentation effects were also observed under these conditions. One explanation of the augmentation effect is that participants learn relationships between the distractor elements, forming configural representations of the contextual pattern. We examined this putative process by training sets of distractors in consistent (AB AB CD CD) or inconsistent configurations (AB AD CD CB). The results suggest equivalent learning of patterns in these two conditions and hence provide evidence against configural learning. 2. Modelling of cue-competition in contextual cuing of visual search Simulation work revealed that a recent model of contextual cuing (Brady & Chun, 2007) predicts a blocking effect using our designs, inconsistent with the observed patterns of data. We have begun work on a new model of contextual cuing which aims to account for the augmentation effects we observed. 3. Cue-competition in sequence learning (4 experiments; 68 experimental sessions) Our examinations of blocking in sequence learning also demonstrated that cue-competition effects do not occur in these incidental learning tasks. In fact, like in our contextual cuing experiments, the consistent trend is towards augmentation. Simulation work with the SRN, a model of sequence learning, has been conducted to show cue-competition predictions in this model. 4. Cue-competition between contextual and sequence cues (2 experiments; 162 experimental sessions) Two experiments examined learning of competing contextual and sequence cues. The results support the findings above and recently published findings in the literature (Jimenez & Vazquez, 2010): cues do not compete for learning in incidental learning tasks. 5. Recognition memory and causal judgements (6 experiments; 141 experimental sessions) Experiments using a human causal learning task found cue-competition effects in both cue recall and cue-outcome contingency recall, suggesting that cue-competition is driven both by impaired encoding of the blocked cue and of the relationship between the cue and the outcome. This encoding failure is in keeping with associative learning accounts. We also observed a difference in responses between the cue-outcome recall and cue-outcome causal judgements, with the latter failing to show cue-competition effects. This finding is in keeping with results in the literature (e.g., Vadillo, Miller & Matute, 2005) and supports the notion that contingency and causality judgements are only loosely related (e.g., Cheng, 1997). In addition, we have used eye tracking technology to explore attentional processes in human causal learning tasks. This work has examined how people process feedback on decisions to update their associations between cues and outcomes. |
| Exploitation Route | Our results will be of interest not only for experimental psychologists working in the area of human learning and memory, but also to researchers working on visual cognition in artificial intelligence. |
| Sectors | Digital/Communication/Information Technologies (including Software),Education |