ERP Waveforms and Memory

When studying infants and children, researchers have documented numerous ERP waveforms that appear to be correlates of underlying neu-rocognitive processes related to memory. As a result, great progress has been made in understanding the neural processes underlying the behavioral manifestations of memory as well as memory development (e.g., Nelson, 1995). An important consideration in examining the impact of maltreatment is the effect it may have on the developing brain (Cic-chetti, 2002b; Cicchetti & Tucker, 1994; DeBellis, 2001). It is partic ularly important to utilize a developmental perspective in considering the effect of maltreatment on ERPs and memory, as there are morphological changes in many ERP waveforms across development that may reflect both maturational changes in the brain and development of the underlying cognitive processes. In addition, it would be quite useful to employ ERPs to investigate environmental influences on brain functioning and development, in particular those brought about by the experience of maltreatment. Although some of the waveforms to be reviewed are not necessarily directly related to memory processing per se, they nonetheless appear to represent neural processing indirectly related to memory. As such, examining these ERP waveforms may provide important insight into the effects of maltreatment on memory.


The P300 (also referred to as the P3b or "classical" P3), perhaps the most extensively studied ERP component, is a positive waveform that peaks between 300 and 900 ms after stimulus presentation, and in adults is typically maximal in electrode sites over the parietal lobe. This waveform is elicited by an infrequent stimulus presented in the context of an oddball paradigm and is believed to represent neural processes involved in context updating (i.e., updating one's representation of the current environment) or revising the contents of recognition or working memory (Donchin, 1981; Rugg, 1995). Source modeling of the P300 has localized its origin specifically to the hippocampal and parietal cortical regions, and possibly the temporal lobe as well (Nakajima, Miyamoto, & Kikuchi, 1994). Studies of individuals with physical lesions in the brain corroborate these findings, with damage to tissue in the temporal-parietal junction inducing a loss of the P300 waveform (e.g., Knight, 1990).

The hallmark of the P300 is its sensitivity to stimulus probability; the amplitude of this waveform reliably increases as stimulus probability decreases (see Duncan-Johnson & Donchin, 1977, for the classic work on this topic). In addition to its relation to stimulus probability, the amplitude of the P300 is also generally greater the more a particular stimulus deviates from the ongoing stream of familiar stimuli (e.g., Fabiani, Karis, & Don-chin, 1990). Distinct from the P300 is another subcomponent referred to as the P3a (an earlier occurring component believed to be generated by an automatic, nonmemory related response to novel stimuli (Knight &

Scabini, 1998). Many studies examining the nature of the P300 itself have shown a multitude of factors that have an effect on the amplitude of the P300, including presentation probability (Duncan-Johnson & Donchin, 1977), stimulus sequence (Duncan-Johnson & Donchin, 1982), stimulus quality, attention, and task relevance of the stimulus (Coles, Smid, Scheffers, & Otten, 1995).

This wide variety of experimental conditions under which the P300 can be elicited has resulted in this waveform being implicated in a range of cognitive processes, and the debate concerning its exact psychological meaning is far from over (Luck, 2005). However, its association with recognition memory through the oddball paradigm has been consistent in the literature, and the general consensus appears to be that the context-updating interpretation of the P300 originally put forth by Donchin (1981), and its relation to updating of working memory (e.g., Rugg, 1995), is the best first approximation to its significance. Aside from the many methodological variations involved in the study of the P300 per se and the disagreement in the literature concerning the exact cognitive process(es) represented by the P300 and the neural generators that give rise to it, there is a vast body of literature supporting its association with updating and/or revising the contents of memory.

Despite the vast literature describing the characteristics and function of the P300 in adults, little work has been done examining the development and nature of this waveform in children. Although some work has shown an auditory P300 produced in the second year of life (e.g., Hoffman & Salapatek, 1981; Hoffman, Salapatek, & Kuskowski, 1981; McIsaac & Polich, 1992), there is some contention that the wave identified as a P300 in this work was not analogous to an adult P300 (see Nelson, 1994, for a critical review). However, early studies by Courchesne (1977, 1978) demonstrated that in children as young as 6 years of age, a P300 elicited by rare, target stimuli in an oddball paradigm was apparent, with a peak latency of approximately 700 ms. The latency of the peak amplitude of the P300 steadily decreased in older children (i.e., the waveform occurred earlier or more quickly), and declined to 400 to 500 ms in young adults. In a study comparing ERP waveforms in adults and a group of 8-year-olds, Thomas and Nelson (1996) demonstrated that the infrequent stimulus in an oddball paradigm elicited a robust P300 in both groups. The findings from this study were consistent with those in the Courchesne (1977, 1978) studies, with the children's P300 waveform broader in shape (less sharply peaked) and occurring later than the P300 observed in the adults.

The P300 waveform appears to develop beginning in middle childhood, with pronounced shifts in characteristics, such as latency, occurring into adulthood. However, it is unclear exactly when (and how) this waveform appears. Thomas and Nelson (1996) demonstrated the P300 in children 8 years old, but the P300 is generally not observed in ERP studies utilizing visual stimuli in infants and young children. A cross-sectional ERP study that included infants, 4-year-old children, and adults demonstrated that while face stimuli elicited a robust P300 in the adults, there was not evidence of a P300 in the 4-year-olds (Scott, Luciana, Wewerka, & Nelson, 2005). However, an earlier study by Nelson and Nugent (1990) demonstrated a waveform with positive peak at approximately 700 ms in a sample of 5-year-olds that was similar in morphology and function to an adult P300. Nelson and colleagues (e.g., Nelson, Thomas, de Haan, & Wewerka, 1998) have speculated that the positive slow wave (PSW; see next section of this chapter) consistently observed in ERP studies of infant memory, and believed to represent the updating of working memory, may be the developmental precursor of the P300. However, to date there are no longitudinal studies of children spanning the period from infancy through middle childhood that could reveal the emergence of the P300 waveform and the morphological changes in ERPs that might accompany this developmental process.

Study of the P300 in the context of maltreatment and memory would allow direct neurofunctional examination of possible memory deficits associated with the experience of maltreatment. Neurobehavioral studies of children who have been maltreated and manifest symptoms of post-traumatic stress disorder (PTSD) report memory deficits in this population (DeBellis, 2001). Thus relevant questions would be whether behavioral evidence of memory deficits in children who have experienced maltreatment is associated with anomalies in the P300 and, more specifically, whether the experience of maltreatment delays the emergence of the P300 waveform. Such research would elucidate the neurofunctional process by which maltreatment may lead to deficits in memory functioning.

ERP studies of school-age maltreated children (Pollak, Cicchetti, Klorman, & Brumaghim, 1997; Pollak, Klorman, Thatcher, & Cicchetti, 2001; Pollak & Tolley-Schell, 2003) have examined the P300 as an index of selective attention as it may be associated with context updating;

however, these investigators have not directly interpreted their findings in the context of memory functioning. An examination of the P300 in children who have experienced maltreatment, particularly the development of the P300 and its relation to memory processes, would be an especially fruitful area of study. Longitudinal examination of the emergence of the P300 from early to middle childhood would provide an unprecedented opportunity to investigate the effects of maltreatment on the development of the brain and neural correlates of memory.

0 0

Post a comment