Developing a memory representation: Do we visualize or do we "verbalize" objects?

Abstract

According to encoding specificity, participants perform better when testing conditions match learning conditions. Interestingly, recent findings in visuo-haptic object identification violate this principle: participants who learned to recognize objects haptically performed just as well when asked to identify objects by sight and by touch. One possible explanation is that participants who explore objects haptically visualize the objects they explore, creating a multisensory memory trace equally accessible to vision and touch. We evaluated this possibility by asking undergraduate participants to learn to recognize novel objects either by sight or by touch. Participants completed sequences of learning trials where they explored each object, and test trials where they recalled the name of each object. During learning trials, some participants were presented with a visual distractor (either a verbal or nonverbal characters) they had to recognize later, while other participants completed a distractor-less control condition. Consistent with past findings, our results violated encoding specificity for participants who learned to recognize objects haptically – this was not modified by the addition of a secondary task. Interestingly however, only the verbal distractors interfered with learning. These results suggest that the creation of memory representations for novel objects involves a verbal code rather than visualization, independently of how objects are initially explored.