تفاوت مربوط به سن در حواس پرتی و جهت گیری مجدد در کار شنوایی

Abstract Behavioral and event-related potential measures of distraction and reorientation were obtained from children (6 years), young (19–24 years) and elderly adults (62–82 years) in an auditory distraction-paradigm. Participants performed a go/nogo duration discrimination task on a sequence of short and long (50–50%) tones. In children, reaction times were longer and discrimination (d′) scores were lower than in adults. Occasionally (15%), the pitch of the presented tones was changed. The task-irrelevant feature variation resulted in longer reaction times and lower d′ scores with no significant differences between the three groups. Task-irrelevant changes affected the N1 amplitude and elicited the mismatch negativity, N2b, P3 and reorienting negativity (RON) sequence of event-related brain potentials. In children, the P3 latency was the same as in young adults. However the RON component was delayed by about 100 ms. In the elderly, P3 and RON were uniformly delayed by about 80 ms compared to young adults. This pattern of results provides evidence that distraction influences different processing stages in the three groups. Restoration of the task-optimal attention set was delayed in children, whereas in the elderly, the triggering of involuntary attention-switching required longer time.

1. Introduction In many everyday situations, maintaining high-level performance in a given task requires focusing on the task-relevant aspects of the environment while disregarding the irrelevant ones. Although goal-directed behavior is primarily governed by top-down control, infrequent unpredictable events are automatically detected (Näätänen, 1992) and can trigger an orienting response (Sokolov, 1963 and Schröger, 1997). Orienting towards unexpected task-irrelevant stimuli is advantageous in an evolutionary sense, because these may carry information that is crucial for survival. However, distraction from one's current task usually leads to temporary deterioration of performance in that task (Escera et al., 2000). Normal functioning of the cognitive system is characterized by a good balance between maintenance of goal-directed behavior and involuntary orientation (passive attention, James, 1890 and Escera et al., 2000). The balance, however, shifts during maturation and aging. Children and elderly adults are more susceptible to distraction than young adults, a fact often attributed to weaker inhibition efficiency related to immature or deteriorated frontal lobe functions (Van der Molen, 2000 and Hasher et al., 2007). However, several processes participate in the distraction-orientation-refocusing cycle and, therefore, changes occurring in the course of human life may affect different functions involved in the interplay between voluntary and passive attention. In the present study, we investigated processes contributing to goal-directed and orienting-related activities in three age-groups (early school-age children, young and elderly adults) with the goal to assess the effects of development and aging on the various processes. 1.1. A three-stage model of distraction Current understanding of the interplay between distraction caused by unexpected events and control processes governing goal-directed behavior can be described in the framework of a three-stage model of distraction ( Escera et al., 2000, Friedman et al., 2001, Näätänen, 1990, Näätänen, 1992, Polich and Criado, 2006, Schröger, 1997 and Schröger and Wolff, 1998b). The first stage of the model features processes, which continuously monitor and “model” the temporal aspects of the sensory environment without the involvement of voluntary control processes ( Schröger, 1997, Näätänen, 1990 and Winkler, in press). Modeling the environment is mainly based on the extraction of regularities from recent stimulation, whereas monitoring is based on the detection of discrepancies between the predictions of the model and incoming stimuli ( Winkler, in press and Winkler et al., 1996). Regularity extraction and deviance-detection is an economic solution to the monitoring problem, as it minimizes the demand on capacity-limited control processes in relatively stable environmental stimulus configurations. Small deviations from the detected regularities usually lead to model-updates, which can be handled within the first processing stage ( Näätänen and Winkler, 1999, Winkler, in press and Winkler et al., 1996). In contrast, major deviations can trigger higher-order processes leading to an involuntary change in the allocation of attention ( Näätänen, 1990, Schröger, 1997 and Escera et al., 1998). That is, gradual changes in the environment occurring over a longer period of time may go unnoticed, whereas the same change occurring rapidly may catch one's attention. The processes of involuntary attention-switching constitute the second stage of the distraction model. Distraction is understood as a transition from a selective attention set which is optimal with respect to performing a given task, to a different, probably suboptimal set (with respect to performing the original task), which might allow for more efficient processing of the distracting task-irrelevant event ( Escera et al., 2000, Polich, 2003 and Schröger et al., 2000). Processes at the third stage of the model are responsible for restoring the optimal attention-set for the task at hand (reorientation), that is, they directly subserve the voluntary re-establishing of the selective attention set appropriate for the primary task ( Munka and Berti, 2006). These processes probably take place only if the task is still relevant at the point when the distracting event has been evaluated. Response execution based on task-relevant information may take place before as well as after the optimal attention-set has been restored. In summary, the first stage can be described as filtering the task-irrelevant stimulation with automatic identification of events that violate the detected sensory regularities. Such events may trigger the involuntary attention-switching mechanisms of the second stage. The third stage encompasses mechanisms that compensate the effects of involuntary orientation to task-irrelevant aspects of the environment by restoring the task-optimal attention-set. 1.2. ERP correlates of the three processing stages The three-stage model receives important support from the analysis of event-related potentials (ERPs) elicited in the oddball paradigms. In the oddball paradigm, occasional irregular stimuli are presented within the sequence of a repeating stimulus. The repeating stimulus is termed the standard, whereas the stimuli violating the repetition are termed deviants. The ERPs elicited in the oddball paradigm reflect many of the processes taking place during the three stages of the distraction-orientation-refocusing cycle. Some of the ERP components are elicited whether or not attention is directed towards the oddball stimulus sequence (corresponding to automatic/involuntary stimulus processing), other ERP components are elicited mainly when participants perform a task related to the oddball sequence. We start our description of the distraction-related ERP components with the ERP components elicited irrespective of the direction of focused attention, which are typically studied in the passive task conditon in which participants perform a task that is not related to the oddball sequence. In the passive oddball paradigm, deviants elicit the modality-specific (auditory, visual and somatosensory) mismatch negativity (MMN), peaking 100–200 ms after the onset of deviance ( Czigler et al., 2002, Näätänen et al., 1978 and Shinozaki et al., 1998). As MMN can be elicited even if participants do not attend the stimuli, it is assumed to reflect a pre-attentive deviance detection process ( Näätänen, 1990 and Sussman et al., 2003b). In terms of the three-stage model, MMN reflects an important process in the first stage, one which detects irregular unattended stimuli. In paradigms utilizing easily discriminable, salient deviants, MMN is often followed by the P3a, a fronto-central positivity peaking at about 300 ms (Friedman et al., 2001) from deviation onset. However, P3a-like activity can also be elicited without a preceding MMN (Rinne et al., 2006), e.g., by rare salient stimuli. P3a is assumed to reflect the activation of an attention-switching mechanism, which is an important step of involuntary orienting of attention (Escera et al., 2000, Friedman et al., 2001, Knight and Scabini, 1998 and Schröger, 1996). Thus P3a would index processes in the second stage of the distraction model. However, there is no general consensus on the precise role of the P3a within the second processing stage (see Dien et al., 2004). In active oddball paradigms (in which participants perform a task related to the stimulus sequence) a number of additional components can be observed. When the oddball sequence is attended, MMN is overlapped/followed by the N2b component peaking around 200 ms (Näätänen and Gaillard, 1983 and Ritter and Ruchkin, 1992). N2b probably reflects a modality-aspecific process: the controlled registration of the occurrence of an infrequent deviant tone (as opposed to its automatic detection reflected by the MMN, see Ritter et al., 1992). The N2b is followed by the P3 components, which typically peak between 300 and 400 ms (for a recent review, see Polich and Criado, 2006). Apart from the already mentioned fronto-central P3a, a centro-parietal, later P3-variant, termed P3b, is elicited when the participant is required to respond to a target deviant. When the discrimination of the target and standards is easy, the P3 component is characterized by a central topography, however, difficult discrimination tasks result in later and more parietal P3b. It has been proposed, that P3b reflects a memory function, which maintains and updates the working-memory representation of the stimulus context on the occurrence of the deviant (Donchin and Coles, 1988, but see Verleger, 1988). When a task-irrelevant deviant distracts the participant from the primary task, a late frontal negativity is elicited 400–600 ms after the onset of the deviation (Schröger et al., 2000, Schröger and Wolff, 1998a and Schröger and Wolff, 1998b). As this component is thought to reflect recovery from distraction, it is referred to as Reorienting Negativity (RON). The processes indexed by RON constitute the third stage of the distraction model. A similar negativity termed late difference negativity (LDN, for a summary, see Cheour et al., 2001) is often found in children in passive oddball paradigms. LDN may be homologous to the reorienting negativity (RON, Wetzel et al., 2006) or reflect higher-order processing of sound change ( Čeponienė et al., 2004). 1.3. A simple paradigm for studying distraction The present study presented a variation of the stimulus paradigm designed by Schröger and Wolff, 1998a and Schröger and Wolff, 1998b. Therefore this paradigm is described in detail in the following. Distraction is often studied using a variation of the oddball paradigm, in which stimuli vary in two features, one task-relevant and the other task-irrelevant. Participants perform a two-alternative choice or Go/NoGo task on every trial based on the task-relevant property of the stimuli (e.g. stimulus duration). The two levels of the task-relevant feature are presented equiprobably. In contrast, the task-irrelevant feature (e.g. pitch) is delivered with unequal probabilities as in an oddball paradigm (e.g., 90% of one level, the standards and 10% of the other level, the deviants). Differences in the processing of deviants and standards are manifested both by ERPs and by indices of task performance. Compared to standards, deviants elicit negativities in the 100–250 ms post-stimulus interval (N1 effect [Näätänen and Picton, 1987 and Jacobsen et al., 2003], MMN, and N2b) followed by the P3a (300–400 ms) and RON (400–600 ms). Responses to deviants are delayed compared to standards, and participants make more mistakes on deviant trials. The notion that RON reflects the recovery of the task-optimal attention set (reorientation) is supported by results of Sussman et al. (2003a), who have shown that a visual cue reliably signaling the occurrence of the deviant feature (the level of the task-irrelevant stimulus feature) eliminates both P3a and RON. This finding can be interpreted as the suppression of the distracting effect of the expected deviant event. Furthermore, larger deviance results in higher P3 and RON amplitudes (Yago et al., 2001). RON exhibits the same topography for visual and auditory stimuli, which suggests that the underlying process is modality-independent (Berti and Schröger, 2001). Escera et al. (2001) suggested that RON is the sum of two subcomponents: one is time-locked to the onset of the deviation, and the other is time-locked to the onset of the target. 1.4. Effects of maturation and aging on distraction-related ERPs and behavioral indices For a summary on developmental changes in attention-related ERP components see Ridderinkhof and van der Stelt (2000). MMN latency and amplitude seems to decrease (or at least not increase) with growing age (Csépe, 1995, Kraus et al., 1993, Kurtzberg et al., 1995 and Lang et al., 1995) in children. It has to be noted that MMN elicitation is often less robust in children when deviance-magnitudes are small (see Räikkönen et al., 2003). Even positive deviant-minus-standard differences in the MMN latency range were found (see Morr et al., 2002 and Maurer et al., 2003), probably due to an overlap of the P3a (Kushnerenko et al., 2002). It is generally found that the latency of P3 decreases with growing age in children up to an age of about 9–10 years (see, e.g. Zenker and Barajas, 1999 and Batty and Taylor, 2002). Distraction-related behavioral and ERP-effects in 5–6 year old children were found to be generally similar to that of in adults in the Schröger-Wolff-paradigm (Wetzel et al., 2004), however, Wetzel et al. (2006) showed that the distraction-related response-delay decreased with age (across 6–9, 10–13, and 19–29 year old groups of participants). These authors found no age-related P3a effects, but the RON amplitude was smaller in 6–9 year old than in 10–13 year old children. RON elicitation was also delayed by about 70 ms in the 10–13 year old group compared to young adults. Wetzel et al. (2006) also found that P3a and RON were elicited even in a passive condition in children but not in adults, which may imply that RON and the LDN are the same component. In school-age children, Gumenyuk et al. (2001) found that deviants (occasional novel environmental sounds embedded in a sequence of pure-tone standards) elicited an early and a late P3 with 200 and 300 ms peak latency, respectively. The early but not the late P3 (probably P3a) exhibited a polarity inversion at the mastoids suggesting that its generators lie in the auditory cortex. P3 was followed by a late negativity (RON), which was larger in younger (7–10 years) than older (11–13 years) children. In the elderly, the MMN amplitude is often lower than in young adults (e.g. Bertoli et al., 2002, Cooper et al., 2006, Czigler et al., 1992, Gaeta et al., 1998, Jääskeläinen et al., 1999 and Woods, 1992; but see Amenedo and Diaz, 1998b, Pekkonen et al., 1996). N2b elicitation is delayed in the elderly (Amenedo and Diaz, 1998a and Amenedo and Diaz, 1998b). It has been shown that P3a and P3b is delayed and elicited with lower amplitude in the elderly (see, e.g. Polich, 1997, Czigler et al., 2006 and Fjell and Walhovd, 2004). Mager et al. (2005) found that the P3 was delayed in middle-aged adults compared to young adults, but there was no difference in the RON-latency. Response-delays to deviants did not differ between the two groups. In an auditory-visual distraction paradigm (odd-even categorization task for visually presented numbers with a synchronous auditory oddball sequence), Andrés et al. (2006) found a larger behavioral distraction-effect for the elderly than for young adults. However, age-related increase of distraction effects is not unequivocal (for a review see Madden and Langley, 2003). The present study investigated distraction and reorientation-related processes as indexed by behavioral and ERP measures in early school-age children, young and elderly adults using an auditory distraction paradigm. The uniqueness of the present approach is that the same paradigm was used in all three age groups, thereby allowing a direct comparison between the response patterns.

5. Conclusions In summary, the present results are consistent with the notion that the automatic filtering of task-irrelevant stimulation is similar in the three groups. Later processes, however, showed an age-dependent pattern. Whereas violations of the detected sensory regularities led rapidly to an involuntary attention-shift in the children and the young adults, this process was delayed in the elderly. On the other hand, the following restoration of the task-optimal attention set commenced with similar speed in both groups of adults, whereas it was delayed in the children. These results suggest that maturation and aging selectively affects different stages in the processing of distracting stimuli, providing evidence against single factor models of cognitive aging (e.g., general slowing, Salthouse, 1996). Since these age-differences were reflected in the ERPs, but not in the reaction times and d′ scores, it can be assumed that processes not revealed by the ERPs also contribute to the deterioration of performance caused by distracting stimuli.