Implicit spatial learning in adult humans

Pratt, Sarah (1999) Implicit spatial learning in adult humans. University of Southampton, Doctoral Thesis.

Record type: Thesis (Doctoral)

Abstract

In this thesis a novel spatial reaction time (SSRT) paradigm was developed to study implicit learning, the acquisition of knowledge largely without earlier conscious effort or explicit awareness of what has been learned. Participants used a computer mouse to track a series of stimuli projected in rapid succession on a two-dimensional visual array. Unlike conventional serial reaction time tasks, the procedure incorporated many-to-one stimulus-response mappings. An artificial grammar was used to structure the sequence of stimulus locations probabilistically. Learning was thus revealed within participants by the decline in reaction times over structured acquisition blocks and an increase during subsequent frequency-matched probe blocks (a negative transfer effect). In experiment one, participants demonstrated significant spatial sequence learning with low accompanying levels of knowledge according to verbal reports. In experiment two, the effect was replicated after fewer acquisition trials, the majority of participants performing at chance levels on subsequent direct tests of sequence prediction and recognition. Experiment three used a between-groups comparison in which experimental participants received the standard SSRT task while controls were exposed to frequency-matched stimulus blocks lacking sequential structure.

The negative transfer effect differentiated experimental and control groups but there were no reliable differences between their performance on measures of recognition, prediction or subjective confidence. Experiment four factorially manipulated task orientation (standard, rule or divided attention) and condition (experimental and control). Although there was a main effect of condition on composite negative transfer scores, there were no effects of task orientation and no interactions. Regardless of orientation, participants performed at chance on two objective tests of learning, an active and a passive version of the recognition task. Overall the novel SSRT paradigm was successful in establishing predominantly implicit sequence learning and offers further scope to extend these initial findings.

This record has no associated files available for download.