Elsevier

Brain and Cognition

Volume 62, Issue 2, November 2006, Pages 113-119
Brain and Cognition

Use of the Hayling task to measure inhibition of prepotent responses in normal aging and Alzheimer’s disease

https://doi.org/10.1016/j.bandc.2006.04.006Get rights and content

Abstract

This study measures the effect of Alzheimer’s disease (AD) and normal aging on the inhibition of prepotent responses. AD patients, normal aged controls, and young subjects were tested with the Hayling task, which measures the ability to inhibit a semantically constrained response, and with the Stroop procedure. AD patients showed a severe deficit in both error rates and response time on the Hayling task. Inhibition was also impaired on the Stroop procedure, both when using raw performance and when using an inhibition score that controlled for reading and naming speed. Normal aged participants showed modest impairment relative to young controls on both tests. Examination of individual performance in AD patients indicated that the impairment was found in most patients on the Hayling test but in only a subgroup of patients on the Stroop test.

Introduction

Alzheimer’s disease (AD) is a degenerative disease that affects a wide range of cognitive functions, including memory, attention, and language. Whereas the majority of clinical accounts of early AD propose that memory functions are affected the earliest, there is increasing evidence to suggest that executive functions are also impaired very early in the disease process and that this impairment could have a significant impact on patients’ autonomy (Amieva et al., 2004, Perry and Hodges, 1999).

The supervisory attentional system (SAS) is a major model of executive functions (Norman and Shallice, 1986, Shallice, 1988). The model is based on the proposition that while most actions are performed automatically, novel, demanding, and conflicting actions require involvement of the SAS. This system is thought to act as an attentional controller by modulating the pattern of activation of action schemas with inhibitory processes. On this view, suppression of irrelevant responses is under the control of the SAS. Cognitive inhibition is defined as a mechanism that actively suppresses distracting information (Hamm & Hasher, 1992) or properties of the distractors, which are in direct competition with the information relevant to the subject’s goals (Tipper, Weaver, & Houghton, 1994). Within this view, inhibition is a flexible mechanism, adaptable to the nature of both the task and distractors.

The distractibility of AD patients and their tendency to make numerous intrusion errors in memory and retrieval tasks suggests that deficient inhibitory processes may contribute largely to their cognitive impairments. Direct measures of inhibition tend to support this view. Sullivan, Faust, and Balota (1995) measured inhibition in this clinical population using the negative priming paradigm. In their procedure, subjects were shown two overlapping objects, one red and the other green and subjects were instructed to name the red object as fast as possible and to ignore the green one. Response time increased when the object serving as a distractor in one trial was used as a target in the very next trial. This negative priming effect suggests that the distractor was first inhibited and thus took longer to activate in the next trial. The authors found impaired inhibition in AD patients using this negative priming paradigm. The Stroop task has also been used to assess inhibition in AD and normal aging. The results on this task are also consistent with an inhibitory breakdown in normal aging and with an amplification of this breakdown in AD (Spieler, Balota, & Faust, 1996). Amieva and collaborators (2002) used a set of computerized tasks in testing the hypothesis that there are different subcomponents of inhibition. These authors investigated whether there was a selective inhibition deficit in this disease or whether it encompassed the whole range of inhibition processes. The results indicated that persons with AD failed to exhibit the negative priming effect and were impaired on the Stroop task. However, they performed normally or at a slightly impaired level on tasks of motor inhibition (go-no go and Stop signal paradigms, respectively). These findings suggest that the inhibition deficit in AD is not a general impairment, but is instead restricted to specific components.

Based on the research described above, there are indications of inhibition deficits in AD patients. However, the extent of these deficits and their relation to the impairments found in normal aging remain to be further understood. Most importantly, both the negative priming and Stroop tasks have been criticized as “pure” measures of inhibition. For example, the interference portion of the Stroop task has been interpreted as reflecting the resolution of a conflict between word reading and color naming and its impairment in AD has been attributed to semantic deficits as well as slowed speed of verbal processing based on a principal components analysis (Bondi et al., 2002). Similarly, a number of authors have argued that the negative priming task indexes episodic retrieval processes (for a review, see Fox, 1995). As a result, other types of inhibition tasks need to be used in AD and additional studies that combine more than one inhibitory task are warranted to confirm that the impaired performance shown by AD patients on these tasks reflects executive inhibition deficits.

Of particular interest in this regard is the use of the Hayling test to study inhibition in frontal lobe injured patients (Burgess & Shallice, 1996). In this task, subjects are asked to complete, as fast as possible, sentences in which the last word is missing. The sentences provide a semantically constrained context such that they are selected to rapidly and automatically induce a particular last word (e.g., ‘Most cats see well at… ?’- night -). In the first condition, subjects are asked to complete the sentences properly, thus reflecting the initiation of a semantically supported automatic response. In the second condition, subjects are asked to refrain from using the automatically activated (or common-sensical) word and to complete the sentence with an entirely unrelated item. This task produces an inhibition of the prepotent response yielded by semantic activation, as subjects have to inhibit the activated word and its semantic associates to perform the task correctly. It has been shown that frontal lobe lesions impede performance on the inhibition section of the Hayling task (Burgess & Shallice, 1996) and that response inhibition in that task is associated with increased activation in a network of left prefrontal areas (Collette et al., 2001). This confirms the executive nature of the Hayling task and its potential for measuring response inhibition in AD. There has been only a single study on the Hayling task in persons with AD and the results indicated a significant impairment in this group (Collette, Van der Linden, & Salmon, 1999). This finding has potential implications at the clinical level, as clinical versions of this test are being developed and distributed (Burgess & Shallice, 1997). Considering the potential application of the Hayling test as a diagnostic tool, additional studies are warranted to assess the degree to which individuals with AD are impaired on the task, the sensitivity to the task at the individual level, the specificity relative to the effect of normal aging, and the relation between this task and classical measures of inhibition. These are elements that were addressed in the present study.

The goal of the present study was to assess inhibition in participants with AD as measured by performance on the Hayling test. Participants were also administered the Stroop task, which is typically used as an index of inhibition in clinical practice, but has been criticized as a poor measure of genuine inhibition in AD.

Semantic deficits and slowing have been proposed to account for the impairment that persons with AD experience in the interference portion of the Stroop task. To control for these potential contributors, we used an inhibition score that takes into account the performance on the naming and reading portions of the Stroop by subtracting from the interference portion of the test the time that the subjects took to perform these portions of the task. For similar reasons, the performance on the first and automatic section of the Hayling test, which measures the ability to complete the test by providing the appropriate word, can be used to examine word finding and/or speed deficits,

An additional goal of this study was to compare the pattern of inhibition deficits in persons with AD to those resulting from normal aging. The cognitive impairments associated with normal aging are frequently interpreted as resulting from a decline in functions associated with the frontal lobes (Moscovitch and Winocur, 1996, West, 1996). From a clinical point of view, a major difficulty in the diagnosis of AD is to distinguish its early manifestations from normal age-related cognitive decline. Thus, it is important to shed light on the specific nature of AD-related inhibition deficit as compared to normal aging. To assess the effect of normal aging we compared the performance of the older healthy participants to that of young participants.

Finally, we were interested in examining the individual pattern of performance in AD patients. AD is often classified as a heterogeneous disease (Habib et al., 1991, Martin et al., 1986, Neary et al., 1986), leading to different cognitive impairments across patients. However, it is crucial to report and qualify this potential heterogeneity because not all components appear to have similar levels of heterogeneity. For example, our previous studies have shown that contrary to the above view, the working memory deficit is remarkably homogeneous even in very mild AD patients (Belleville et al., 1996, Belleville et al., 2003). In this context, examination of individual profiles is likely to provide important information regarding the pervasiveness of the inhibition deficit in AD. It may also shed light on the way inhibition tasks should be used in clinical practice. Tasks on which impairment is highly homogeneous are likely to be powerful diagnostic markers of the disease. In turn, tasks on which performance has a marked heterogeneity are likely to be qualifiers of differential impairment and potential markers for patient subtyping and for the development of differential modes of intervention. Thus, examination of individual differences was conducted in this study by deriving individual Z-scores in patients with AD.

Section snippets

Participants

Thirty-six participants took part in the experiment: 12 young, 12 old, and 12 patients suffering from AD. The young adults (6 males and 6 females) were between 19 and 30 years of age (M = 22.0), and had a mean of 13.8 years of education (SD = 1.7). The older adults (4 males and 8 females) were between 68 and 83 years of age (M = 72.7) and had a mean of 11.0 years of education (SD = 2.0). The AD patients were between 64 and 85 years of age (M = 72.5; SD = 5.99) and had an average of 10.1 years of education (

Hayling test

Two dependent measures were used in the Hayling test: response latency and error rate. The median (in seconds) was calculated for each condition and participant to obtain a response latency score. An average response latency score was then computed based on all responses including errors. In the inhibition condition, responses were scored according to the criteria proposed by Burgess and Shallice (1996). This was done to allow direct comparison with Burgess & Shallice’s data collected with

Discussion

The Hayling test was used in AD patients, normal aged persons and young participants to measure the inhibition of prepotent responses. In addition, the Stroop task, a classical test of inhibition was used as a comparison. At the group level, our findings strongly support the presence of severe inhibition deficits in groups of AD participants when considering performance on both tasks. Indeed, AD patients were slower and made more errors than elderly adults on the inhibition section of the

Acknowledgments

S.B. received support by the Fonds de la Recherche en Santé du Québec and by the Canadian Institute for Health Research and N.R. received support by the Fonds pour la Formation de Chercheurs et l’Aide à la Recherche and the Fonds de Développement Scientifique of the Belgian government. Special thanks to Édith Ménard and Paule Rouleau for their help with the testing and data analysis and to Janet Boseovski and Lyssa Gagnon for revising the text.

References (32)

  • J. Chatelois et al.

    Batterie informatisée d’évaluation de la mémoire inspirée de l’approche cognitive

    Canadian Psychology/Psychologie canadienne

    (1993)
  • E. Fox

    Negative priming from ignored distractors in visual selection: a reviews

    Psychonomic Bulletin & Review

    (1995)
  • Gérard, M. (1983). Contribution à l’évaluation de la détérioration mentale chez l’adulte à l’aide du test de...
  • C.H. Golden

    Identification of brain disorders by the Stroop color and word test

    Journal of Clinical Psychology

    (1976)
  • M. Habib et al.

    Démences et syndromes démentiels: Approche neuropsychologique

    (1991)
  • V.P. Hamm et al.

    Age and the availability of interferences

    Psychology and Aging

    (1992)
  • Cited by (106)

    • Swedish Version of the Hayling Test: Clinical Utility in Frontotemporal Dementia Syndromes

      2019, Journal of the International Neuropsychological Society
    View all citing articles on Scopus
    View full text