[Aging and cognitive slowing: example of attentional processes--evaluation procedures and related questions]

Slowing is generally associated to ageing. It appears in motor's functions and in cognitive tasks. What is the real nature of this slowing? Is it a general slowing concerning every cognitive processes with the same scale or is this slowing specific of only processes. Or, at least, is it of different magnitude for each cognitive processes? The aim of this paper is to present the state of this debate from results obtained in studies orientated toward attentional processes. Attention allows us to adapt oneself to environment that require in one hand selective mechanisms for pertinence events and in other hand inhibitory mechanisms for interferences. To evaluate these mechanisms priming and cueing procedures are used. Using primes (semantic or conceptual) results in shorter reaction time than for conditions without prime. In some experimental conditions, negative primes can be observed which results for longer reaction time. In these procedures, we need to be careful to the SOA's value (Stimulus Onset Asynchrony) (it represents the time between the end of the prime or cue presentation and the beginning of the stimulus presentation). The longer is the SOA, the shorter is the reaction time increase with a SOA between 200 ms and 400 ms. In these procedures, we now have to understand in the field of information processing what causes this reaction time modification. In other words, increase of reaction time with SOA could be explain increase of all stages of treatment or may be also the consequence of some abolition's of stages? In attentional procedures, we have to consider the automatic or controlled nature of cognitive processes. In target research tasks that implies selective attention, several authors have showed a distinction between automatic and controlled processing. Time to detect prompts increase generally with the distractor number excepted when the prompt is prominent for the subject (because of physical or emotional characteristics). In this second case, the pop out effect, reaction time does not depend from the number of distractors and this effect performs to be supervised by automatic processes. By contrast, reaction time variation is considered to be linked with controlled processes activation. What's about ageing? Myerson et al. (1992) don't show reaction time distinction between young and old in their meta analyse which really results from lexical decision tasks from 1980 to 1990. If you now consider the SOA values, results are less clear and contradictions existing between studies. Using negative priming procedures implies different results too, but one is main: it seems that elders take as much time to denote a prompt in conditions when prompt was distractor compared to neutral condition, by contrast to younger who take more time to denote the prompt where it was distractor compared to condition where it was not. Ageing effects on inhibition intensity and length could be expanded. Nissen and Corkin (1985) showed a facilitator effect with cueing in young and in old people, and this appears more important in old people. Hartley et al. (1990) show this results too and control the SOA: the facilitation size effect seems to depend from the SOA size. On top of that, some authors show the presence of the popout effect in elders but a reaction time lengthening when conditions imply controlled processing (visual or memory research tasks). In summary, the most studies show a slowing in old people whatever attentional tasks but the lengthening size doesn't vary systematically with conditions: facilitatory effects and automatic processing are possible but inhibitory processing are more probably impaired. Several hypotheses can explain these results. Some authors think there is a general slowing factor that slows with age. For others slowing depends from tasks, and mainly from several implied processing. Nevertheless, no one study show a regular relation between reaction time and priming or cueing effects. Few results, which we've presented in this article, are really compatible with the general slowing theory. Finally, McDowd and Birren (1990), or Hasher and Zacks (1988) suppose that there is an incapacity to ignore non pertinent information with ageing, but there are limit in this hypothesis too. In summary, we don't know at this time the real nature of slowing with age but experimental values seem to be in favour with slowing, which significance depends from cognitive processes. We now have to describe precisely slowing in old people, and more exactly with exact timing between prime and target could be a beginning of task.