[Inhibition and resource capacity during normal aging: a confrontation of the dorsal-ventral and frontal models in a modified version of negative priming].

BACKGROUND: Over the past decades, cognitive psychology contribution to our understanding of aging relies on two major perspectives, focusing on the selective impact of age on either cognitive multiple-systems or global factors of cognition: slowing, working memory and inhibition. In the latter, reduction in inhibitory control during aging (in its access, deletion or restraint functions) is associated with poorer performance on a variety of tasks referring to memory, comprehension or language [Hasher, Zacks and May (16)]. The attractiveness of inhibition as an explanatory factor results in part in the absence of negative priming during aging. Negative priming refers to the slow down of latencies when individuals have to respond to recently ignored informations, compared to unrelated informations. The dissociation, between a preserved location negative priming and an absence of identity negative priming during aging, supports the dorsal-ventral model of inhibition which suggests that spatial and identity inhibition are supported by different and independent visual pathways. An alternative model, directly at odds, is that inhibitory mechanisms are supported by the frontal lobe. In this perspective, inhibition is not a central process responsible for the control of working memory contents, but an automatic and local mechanism whose triggering depends on controlled attention. Therefore, working memory drives efficient inhibition by sustaining task instructions and appropriate responses throughout task execution. This hypothesis is consistent with Houghton and Tipper's (17) architecture of selective attention. According to the authors, the presence or absence of automatic inhibition is very closely linked to a Match/Mismatch field whose function is to compare the present stimulus to an internal self-generated internal template. When an information fails to match the subject's current goals, the match/mismatch field causes an automatic inhibitory imbalance which reduces the to-be-ignored properties' responsiveness. In contrast, information matching subjects' goal is enhanced through an automatic excitatory imbalance. The accurate functioning of the Match/Mismatch field requires efficient executive functioning responsible for the uphold of goals and correct responses. In the case of negative priming, manipulating the efficiency of working memory is of interest as it should affect the triggering of slowing, ie, an indirect inhibitory deficit, when the task is resource demanding [Conwayet al. (6)]. Moreover, if inhibition, as reflected by negative priming, is mediated by individual resource capacity, then NP should disappear during aging only when individuals are engaged in a resource-demanding task.

OBJECTIVES: To address this issue, we examine whether cognitive control load in a gender decision task contributed to the presence or absence of NP during aging. According to the dorsal-ventral model, task complexity should not have any impact on performance, since gender decision task relies on a conceptual analysis of information. In turn, the frontal model predicts that age differences in performance profile will only differ when individual resource capacity is overloaded.

DESIGN OF THE STUDY: Sixty-four participants (32 young and 32 older adults) performed a gender categorisation task through two experiments. Trials involved two stimuli presented successively at the same location. A word served as a prime and a word as a target. Both prime and target could be male or female. When prime and target matched on gender, we talked about VALID pairs (or compatible). When prime and target mismatched on the manipulated features, we talked about INVALID pairs (or incompatible). Participants' task was to identify the gender of the target. They were explicitly instructed not to respond to primes but to read them silently. Our interest was in response latencies for valid versus invalid pairs. We manipulated task complexity by the absence (experiment 1) or presence (experiment 2) of a distractor during probe trials.

RESULTS: For younger adults, primes presented before targets gave rise to behavioural costs when pairs were mapped to the same response compared to pairs that were mapped to opposite ones. This slowing, called Negative Compatibility Effect (NCE), was independent of the presence or absence of a distractor. NCE was reliable for the elderly patients only under condition of no interfering information during the probe trial: pattern of performance of older participants was identical to that of young adults (experiment 1). This effect disappeared as task complexity increased (experiment 2).

DISCUSSION: This result suggests that NCE triggering is dependant on the amount of cognitive control engaged by the task, and therefore that the ability to inhibit irrelevant information is secondary to a general capacity of the working memory.

CONCLUSION: The implications of our data are consistent with the level of processing account, as well as the recent neuroimaging contributions which suggest, for example, the involvement of the dorso-lateral prefrontal cortex (sensitive to aging) when task demands are high, and a ventro-lateral prefrontal implication when demands are low [see Eenshuistra et al. for a review (10)].