Research report
Primary task demands modulate P3a amplitude

https://doi.org/10.1016/S0926-6410(99)00044-0Get rights and content

Abstract

Auditory event-related brain potentials (ERPs) were recorded from 10 subjects in two different conditions: (l) subjects were required to reorder five visually presented letters in order to form a word and provide a verbal response (task condition); (2) subjects were presented with a control stimulus with the same physical characteristics as the experimental stimulus, but containing just one type of letter (i.e., AAAAA). Subjects had to verbally respond to such stimuli by saying “A” (control condition). Tones of 1000 Hz (standard) and 1050 Hz (deviant) were also presented to the subjects in a 85%–15% probability paradigm 2 s before, during and 8 s after the presentation of the visual stimuli. Recordings were obtained from Fpz, Fz, Cz and Pz vs. linked ears. Auditory ERPs to the auditory stimuli after the presentation of the visual letter string and during the performance of the task were averaged for the standard and deviant tones in both conditions. Only correct responses were considered for the averages. The N100 was affected by stimulus type (standard vs. deviant) but not by condition (task vs. control); however, larger P3a waves were observed during the control than during the task condition. No significant differences between conditions were observed in the mismatch negativity (MMN) latency range. These results suggest that primary task demands modulate involuntary attention processing.

Introduction

The P300 component of the event-related brain potentials (ERPs) reflects fundamental cognitive processes [5,14]. In general, this component is obtained using the so-called oddball paradigm, wherein two stimuli are presented in random order, one of which occurs less frequently than the other. The participant is required to discriminate the infrequent stimuli (target) from the frequent one (standard) by noting the occurrence of the target (mentally counting them or by pressing a button, for example). The P300 elicited by the target in such a task is a large, positive wave that is largest over parietal electrode sites and occurs at about 300 ms after stimulus onset in normal adults and is more specifically known as P3b. Squires et al. [20] described another positive wave, evoked by occasional shifts in either pitch or loudness in an ongoing train of tone bursts occurring in both attended and ignore conditions, with a frontocentral distribution [19] and called it P3a. Later on, a modification of the oddball paradigm was used in which a third (novel), infrequent nontarget stimulus was presented in addition to the sequence of standard and target stimuli. Using this method, Courchesne et al. [4] described a frontocentrally distributed P300 to novel stimuli, and also labeled it as P3a [11,15]. In sum, a P3a may be produced by several types of nontarget stimuli: infrequent stimuli in a passive condition, physically novel stimuli, or infrequent nontarget stimuli in a three-or-more-item oddball task.

According to Picton [14], the P3a may be distinguished from the P3b by the fact that P3a is not affected by whether the subject is attending the stimuli or not. In a similar fashion, Jocoy et al. [9] considered that the P3b reflects voluntary focused attention, whereas the P3a reflects involuntary attentional capture with automatic processing of novel stimuli. Virtually, all reliable reports of the P3a phenomenon have used stimuli that produce a frontocentral “alerting” response that most likely originates from neural sources related to initial attention allocation [10]. Näätänen [12] distinguished the P3a and the P3b by considering that the P3a is a reflection of an attentional switch produced by the mismatch between a stimulus presented which deviates from the passively formed neuronal trace of another stimulus. The P3b, on the other hand, reflects the match between the stimulus and a voluntarily maintained attentional trace.

In order to evaluate the factors that may modify the P3a component, Katayama and Polich [10] used a three-stimulus oddball paradigm in which participants responded only to targets during four different experimental conditions defined by a combination of two levels of target/standard discrimination difficulty (easy vs. difficult) and two levels of magnitude difference between the physical characteristics of the nontarget/standard (large vs. small). Thus, the four experimental conditions were: easy/small, easy/large, difficult/small and difficult/large. They observed that, in all conditions, target tones elicited a parietal P300, which was affected only by the target/standard discrimination ease (easy vs. difficult). Nontargets in the easy/large and difficult/small conditions elicited similar ERPs to the standard. However, nontarget stimuli in the difficult/large condition elicited an earlier and more anterior P300 (P3a), suggesting that P300 to targets are not influenced by the nontarget stimulus configuration. P300 to nontargets, however, are determined directly by the stimulus context. Hence the manipulation of nontarget/standard tone stimulus context produced a task environment in which the nontarget stimulus quickly engaged focal attention in a similar manner to that observed previously for highly “novel” stimuli. In another study, Comerchero and Polich [3], using also a three stimulus oddball paradigm, manipulated the stimulus characteristics of an infrequent nontarget, such that its perceptual distinctiveness from the target was low or high. This was done in both auditory and visual modalities, separately. In both conditions, the amplitude of the P300 to targets was larger than to nontargets, but only at the parietal electrode. In contrast, nontarget P3a amplitudes were larger and earlier than target P300s over frontal/central electrodes. From these results, the authors considered that perceptual distinctiveness of the eliciting stimulus contributes to the amplitude of P3a.

Attention capacity theories originated from the observation of an interference between simultaneously executed tasks. In such theories, it is assumed that psychological processes require the use of several specific structures, as well as the engagement of a certain amount of resources. It is further assumed that these resources are limited. P300 studies during dual task paradigms show that, as the perceptual demands of the secondary task increase, the amplitude of the P3b evoked by the detection of targets in the primary task decreases [21]. Although it has been described that the amplitude of the P3b is relatively unaffected by increasing the motor demands of the secondary task [7], Schubert et al. [18] recently described that the amplitude of the P3b to an auditory target stimulus was reduced during a forced task requiring a precise movement, whereas no concurrent motor task effect was observed on the P3a.

Ahlo et al. [1] examined the influence of processing load during intermodal attention tasks. Frequent 1000 Hz tones (standards) were presented with an 80% probability. Infrequent tones of 1050 Hz (deviants) and 1500 Hz (DEVIANTS) were presented with a 10% probability each. There were three auditory attention conditions: (a) The “difficult auditory discrimination” condition, in which subjects pressed a button in response to deviants; (b) the “easy auditory discrimination” condition, where subjects responded to DEVIANTS; and (c) the “auditory targets alone” condition, in which auditory standards and deviants were omitted from the stimulus sequence and only infrequent DEVIANTS were left in the auditory modality, subjects had to respond to all tones. In the difficult auditory condition, the negativity to nontarget DEVIANTS was followed by a frontocentral positivity peaking between 250 and 300 ms (Fig. 3 of Ref. [1]), which was interpreted either as a positive deflection (Pda [22]) or a P3a [12]. In a more recent paper, Alho et al. [2] analyzed the effect of ignored auditory stimuli during a visual task. Auditory stimuli consisted of a 600-Hz standard tone or a 700-Hz deviant tone occurring randomly at a probability of 0.2. When deviant tones preceded the visual target, they elicited a mismatch negativity (MMN) and a P3a which also caused an increase in reaction times and error rates in the visual task. These results indicated involuntary attention to an auditory stimulus change. These effects were observed even when the deviant tones occurred simultaneously with a visual warning stimulus introduced to keep attention focused on the visual task. Similar results have been reported by Schroeger and Woll [17]. Thus, there is a general consensus that in dual task paradigms, deviants in an ignored oddball auditory sequence elicit both MMN and P3a components indicating involuntary attention to an auditory stimulus change and that there is a concomitant decrease in performance accuracy in the visual task. In a study of acute effects of ethanol on involuntary attention shifting, Jaaskelainen et al. [8] reported the suppression of the P3a with a very small dose of alcohol. The above mentioned results suggest that the P3a is a very sensitive index of involuntary attention and that it may be used for the evaluation of primary task demands. Primary task demands have been previously evaluated by the “probe” method described by Papanicolau and Johnstone [13]. In this technique, task-irrelevant stimuli (probes) are presented during the performance of different tasks. ERPs to auditory probes were significantly reduced over the left hemisphere during arithmetic calculations, but smaller over the right hemisphere during a visuospatial task. Thus, differential task demands may modulate the amplitude of different ERP components. In this paper, we were interested in the effect of different load demands over the ERPs obtained to ignored stimuli. Subjects were engaged in two different verbal tasks, and nonrelevant auditory stimulation was simultaneously presented. Two different tones were used to evaluate the effect of the task on the evoked responses to both standard and deviant stimuli.

Section snippets

Subjects

Subjects were 10 right-handed volunteers (22–34 years of age), five of which were females. All subjects had a normal EEG and no history of neurological problems.

Tasks

Two different types of visually presented five-letter strings were presented during 1 s on a videomonitor in random order and with equal probability (Fig. 1), namely: (1) Experimental or task condition: five uppercase letters, to be reordered by the subject to form a word (i.e., ABTEL to form TABLE). After 8 s after stimulus

Number of trials

After the rejection of incorrect responses (50%–70%) the number of segments to average was between 450 and 700 for the standard tones and between 80 and 150 for the deviant tones during the task condition. The same number of segments for standard and deviant tones were randomly selected during the control condition.

ERPs to standard tones (Fig. 2) were characterized by small P1s (mean latency 45 ms and mean amplitude 0.58 μV at Cz), N1s (mean latency 100 ms and mean amplitude −0.60 μV at Cz), P2s

Discussion

The ERPs recorded have the same overall characteristics that have been reported in the literature. An N100 was clearly observed in the ERPs to standard stimuli in both conditions. ERPs to deviant stimuli in both conditions showed a larger and wider negativity which may have contained both an N1 and an MMN, which makes it impossible to measure N1s and MMNs separately. The ERPs to deviant stimuli showed that this negative wave was followed by a positive wave peaking between 230 and 300 ms, maximal

Acknowledgements

The authors want to acknowledge the technical assistance provided by Engineer Miguel Tomás Rodríguez Espinoza and Ms. Pilar Galarza for documentation aid. They want also to acknowledge Dr. Maritza Rivera for the review of the English version of the manuscript. This project was partially supported by grant IN209998 from DGAPA, UNAM.

Cited by (0)

View full text