Academic journal article Psychonomic Bulletin & Review

Sensorimotor Simulations Underlie Conceptual Representations: Modality-Specific Effects of Prior Activation

Academic journal article Psychonomic Bulletin & Review

Sensorimotor Simulations Underlie Conceptual Representations: Modality-Specific Effects of Prior Activation

Article excerpt

According to the perceptual symbols theory (Barsalou, 1999), sensorimotor simulations underlie the representation of concepts. Simulations are componential in the sense that they vary with the context in which the concept is presented. In the present study, we investigated whether representations are affected by recent experiences with a concept. Concept names (e.g., APPLE) were presented twice in a property verification task with a different property on each occasion. The two properties were either from the same perceptual modality (e.g., green, shiny) or from different modalities (e.g., tart, shiny). All stimuli were words. There was a lag of several intervening trials between the first and second presentation. Verification times and error rates for the second presentation of the concept were higher if the properties were from different modalities than if they were from the same modality.

In recent years, researchers have increasingly argued that cognition is based on interactions between sensorimotor systems and the physical world. According to the perceptual symbols theory (Barsalou, 1999), conceptual knowledge is not amodal but instead is grounded in physical experiences. The basic assumption of the theory is that a concept is represented by running a simulation of an actual experience (i.e., of perception and/or action) with that concept. For example, to represent the concept CHAIR, neural systems for vision, action, touch, and emotion partially produce the experience of a particular chair. Behavioral studies increasingly support the view that sensorimotor systems play a critical role in cognition (e.g., Glenberg & Kaschak, 2002; Pecher, Zeelenberg, & Barsalou, 2003; Solomon & Barsalou, 2001; Spivey, Tyler, Richardson, & Young, 2000; Stanfield & Zwaan, 2001; Zwaan, Stanfield, & Yaxley, 2002, Zwaan & Yaxley, 2003), as does neural evidence (e.g., Martin & Chao, 2001; Martin, Ungerleider, & Haxby, 2000; Pulvermüller, 1999).

An important assumption of the perceptual symbols theory (Barsalou, 1999) is that perceptual simulations are componential and dynamic. Rather than act as rigid symbols, perceptual symbols vary widely in their activation patterns, depending on the context in which the concept is encountered (for earlier views on the flexibility of representations, see Anderson et al., 1976; Barsalou, 1982, 1993; Greenspan, 1986; McKoon & Ratcliff, 1988; Zeelenberg, Pecher, Shiffrin, & Raaijmakers, 2003). For example, the simulation of APPLE will be different for someone painting a still life than for someone cooking a meal. The simulation will likely include the shape (round) and color (green) of an APPLE for someone who is painting, whereas the simulation will likely include the texture (crunchy) and taste (tart) of an APPLE for someone who is cooking. Thus, simulations may vary in the extent to which information from various modalities is included.

Recently, we have obtained evidence that modalities play an important role in the composition of simulations (Pecher et al., 2003). Using properties from six modalities (vision, audition, taste, smell, touch, and action), we found that there was a modality switch cost in a property verification task in which a different concept was presented on every trial (i.e., concept names were not repeated). For example, subjects might verify the auditory property loud for BLENDER. On the preceding trial, subjects verified a property for a different concept that was either from the same modality (e.g., LEAVES-rustling) or from a different modality (e.g., CRANBERRIES-tart). Responses were faster and more accurate on sawe-modality trials than on different-modality trials. This effect could not be explained by associative relations between the properties. Rather, the effect is explained by assuming that the simulations are executed by the same systems that are used for perception, and different systems are used for different modalities. …

Search by... Author
Show... All Results Primary Sources Peer-reviewed

Oops!

An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.