Habituation: an event-related potential and dipole source analysis study

Abstract

The goal of this study was to investigate habituation processes in the brain, and in particular, to identify the brain structures involved in these processes. Therefore, event-related potentials (ERPs) were recorded in response to a series of repeated, task-irrelevant, salient stimuli presented against the background of a cognitive task which required the subjects to attend to a fixation sign on a TV screen. Stimuli were presented randomly in one of four possible quadrants of the TV screen. Dipole analyses were performed in order to localize the source of stimulus repetition effects (habituation). As expected, there was a decrease in amplitude of the N1 and P3 components as a function of stimulus repetition; habituation of N1 occurred faster than that of the P3 component. It is suggested from the results of dipole analyses that the N1 effect concerns diffuse cortical activation. The P3 habituation effect seemed to involve the temporal cortex especially.

Introduction

Decrease of response amplitude as a function of stimulus repetition is one of the basic properties of the Orienting Response (OR) (Siddle et al., 1983). This property has been demonstrated for the skin conductance reaction (SCR). Some components of the event-related potential (ERP) have also been found to decrease to repeatedly presented stimuli (Rust, 1977, Wastel and Kleinman, 1980, Barry et al., 1992, Geisler and Polich, 1994, Kazmerski and Friedman, 1995, Cycowicz et al., 1996, Romero and Polich, 1996, Cycowicz and Friedman, 1997, Friedman et al., 1998, Ravden and Polich, 1998). Verbaten et al. (1986) and Kenemans et al. (1989) showed that the N1 and P3 components particularly decrease in amplitude as a function of stimulus repetition. However, there appears to be a pronounced difference in speed of habituation between the two ERP components. Verbaten et al. (1986) found that the decrease in the amplitude of the vertex N1 was much faster than that of P3. Furthermore, the latter authors reported that the vertex N1 and the skin conductance reaction habituated equally fast under non-signal (=task-irrelevant) conditions. These data of Verbaten et al. (1986) suggest that the vertex N1 is involved in a more crude type of stimulus processing than the P3 vertex. According to the most influential current hypothesis about the psychological significance of P3, in the context updating hypothesis (Donchin, 1981, Donchin and Coles, 1988), P3 reflects the updating of a proper representation of the environment in memory. It has to be acknowledged here, however, that alternative interpretations of P3 exist, such as the hypothesis of Verleger (1988) (see also Verleger, 1998), in which P3 is assumed to be elicited by stimuli that close a (meaningful) context (context closure instead of context updating).

The first aim of the present study was to try to replicate the fast habituating N1 effect and furthermore, to determine the neural sources involved in this effect by using the BESA dipole approach (Scherg, 1990, Scherg and Picton, 1991).

The second purpose of this study concerned the P3 component and its habituation. In current theories about visual information processing, posterior-cortical structures play an important role in that a dorsal pathway to the parietal cortex concerned with spatial information is distinguished from a ventral pathway to the inferior temporal cortex concerned with non-spatial information (e.g. Mishkin et al., 1983, Livingstone and Hubel, 1988, Zeki and Ship, 1988, Desimone and Ungerleider, 1989, Kandel, 1991). It may be assumed that both spatial and non-spatial stimulus information are part of the ‘neuronal model’ (Sokolov, 1963), are updated in memory (Donchin, 1981, Donchin and Coles, 1988; see however, Verleger, 1988) during repeated stimulus presentation, and are reflected in the P3 component. If the non-spatial properties (identity) of the stimuli are constant, and the location of the stimuli varies randomly, it may be expected that habituation will occur faster in brain structures involved in identity processing, i.e. along the ventral route (e.g. the inferior temporal cortex) than in structures (sources) involved in location processing (the dorsal route). This question was investigated by applying the BESA approach to ordinally averaged ERPs (see also Ritter et al., 1968).