Stroop interference and negative priming in patients with multiple sclerosis
Introduction
A number of studies have suggested that multiple sclerosis (MS) patients have attentional problems. Some researchers [1], [4] use the term attention in its broadest sense, considering its role generally in various capacity demanding tasks. Others [22], [34] have distinguished between the different types of attention, such as sustained, divided or selective attention. For example, MS patients have been found to be impaired on the Stroop colour/word selective attention task [22], [36]. In this task, participants can be asked to name the colour of the ink that a series of words is written in. When word and colour conflict (e.g. the word red written in blue ink), colour naming latencies are longer relative to a neutral/control condition (e.g. patches of colour, or coloured strings of XXXs). MS patients show increased interference effects when compared with control participants. It would seem fair to say, however, that where specific attentional processes have been looked at, the precise mechanisms which may be impaired have not been investigated. In the last 10 years or so, theoretical accounts of attention have suggested that selection may involve both excitation of the target [13] and inhibition/suppression of distractor representations [2], [17], [18], [40]. The experiment reported in this paper represents an initial attempt to assess whether a breakdown in inhibitory processes can explain the apparent selective attention deficit shown by MS patients when tested on the Stroop task.
Interestingly, deficits in frontal lobe functioning have been found in some MS patients [3], [8], [15], [28] using tasks designed to tap executive functions (e.g. Wisconsin Card Sorting task), and the frontal lobes have themselves been implicated in inhibitory control of interference [12]. Impairments in inhibitory processes have also been implicated in a number of different populations (e.g. people with schizophrenia [5], [6]; people with Parkinson’s disease [7]; and the elderly in general [16], though see also [41]), and here the Stroop task has often been used to demonstrate such deficits. There is an underlying assumption in the clinical literature that this task in itself reveals impaired inhibitory processes. The interpretation usually given to the increased interference shown by these populations is that individuals are less able to suppress the distracting word. Yet the Stroop task does not actually give direct evidence of inhibition. The assumption that inhibitory processes are impaired may or may not be correct. Some of the original explanations of interference in the Stroop test made no reference to inhibition (see [27] for review) and more recent theoretical accounts of selective attention in the Stroop task can include excitatory, inhibitory and other relevant components, such as cognitive units which represent task demands [9]. Increased interference effects in this selective attention task could, therefore, theoretically be due also to an impairment in excitatory processes [21], or perhaps to a problem in maintaining the goal of the task (e.g. that it is colour rather than word naming that is required [10], [43]). Therefore, it is important to demonstrate evidence for impaired inhibitory processes in tandem with increased Stroop interference.
The theoretical research on selective attention referred to above uses what is referred to as the “negative priming paradigm” to infer inhibition. Tipper et al. [39] used a simple and modified card version of the Stroop task which included a negative priming condition, which was referred to as “ignored repeated” condition. Here, the colour on each conflict trial matched the ignored word on the previous trial. If participants inhibit word representations on one trial, then they should be delayed in retrieving the next colour. The “ignored repeated” conditions should, therefore, have overall longer naming latencies than the standard conflict conditions (the negative priming effect). It must be noted, that others [29], [31] have challenged the inhibitory interpretation of this effect.
In the following exploratory experiment, we use the card version of the Stroop test for ease of presentation with the patient group, and we include the neutral, Stroop conflict and ignored repeated conditions used by Tipper et al. [39]. Comparisons of neutral condition naming times for patients and matched controls need to be carried out. There can be specific difficulties in interpreting data from clinical or elderly populations when baseline performance differs [25], [42], and one approach is to compare the proportionate increase in interference [6], [39] (i.e. increase in interference relative to baseline naming times). If increased interference in the Stroop conflict condition is evident for patients, then we can examine whether this might be due to a breakdown in inhibitory processes. Verhaeghen and De Meersman [42] have noted that the literature on negative priming in the elderly has tended to concentrate on the presence or absence of negative priming; the presence of negative priming is inconsistent with a complete breakdown of inhibition. The first step, therefore, would be to examine whether MS patients show any evidence for inhibitory processing; are response times (RTs) longer in the ignored repeated condition than the Stroop conflict condition? If MS patients exhibit negative priming, the second step would then be to examine whether negative priming effects are weaker for MS patients than for controls, which might be indicative of partially impaired inhibitory processes. Finally, it is of interest to include also congruent conditions, (where colour and printed colour name match), since facilitation in this condition relative to the baseline can be an indication of response strategies [6]. Do both patients and controls show faster naming times in congruent conditions?
MS patients experience depression during their illness [1], [23]. Hasher and Zacks [16] suggested that depression may influence tasks requiring attentional capacity, and it has frequently been suggested that depression may underlie many of the cognitive deficits observed in these patients. For example, Arnett et al. [1] found some indication of poorer performance for depressed mood MS patients on certain cognitive tasks which required attention (though see [11], [23]). Filippi et al. [14] also found that depression in MS patients was related to frontal lobe function, as measured by the Wisconsin Card Sorting Test and verbal fluency. We include a self-report measure of depression in the present study, and we will check the relationship between depression and interference scores.
Section snippets
Participants
Twenty-one MS sufferers at a self-help centre in Shrewsbury, Shropshire responded to a notice which asked for volunteers to take part in a study examining cognitive processes in MS and non-MS sufferers. One MS participant was subsequently excluded because it was clear that the experimental instructions had not been understood. Of 30 control volunteers, 20 were selected on the basis that they most closely matched the MS participants on a number of variables, as detailed below. Control
Results
For each participant, the mean of the naming times (RTs) for the three cards in each condition was calculated. Table 1 shows the mean of these means, collapsed across participants, along with the percentage error rate.
The RT data were analysed using a split plot ANOVA. The main effect of group was significant (F(1,38)=30, P<0.001). The effect of condition was also significant (F(3,114)=149.67, P<0.001), as was the interaction between the two factors (F(3,114)=12.68, P<0.001). Simple effects,
Discussion
Both MS patients and control participants took longer time to name colours in the Stroop conflict condition than in the neutral condition, and took longer time to name colours in the ignored repeated condition than in the conflict condition. Although RTs were fastest in the congruent condition for both groups, this condition did not differ significantly from the neutral condition. It has been reported that facilitation in the congruent condition is less easy to detect than interference in the
Acknowledgements
We thank two anonymous reviewers for the points concerning heterogeneity of the MS sample.
References (43)
- et al.
What is the nature of increased Stroop interference in schizophrenia
Acta Psychologica
(1999) - et al.
Stroop interference and disorders of selective attention
Neuropsychologia
(1996) - et al.
Memory dissociation and metamemory in multiple sclerosis
Neuropsychologia
(1999) - et al.
Mechanisms of attention: a developmental study
Journal of Experimental Child Psychology
(1989) - et al.
Depressed mood in multiple sclerosis: relationship to capacity-demanding memory and attentional functioning
Neuropsychology
(1999) - et al.
Externally cued and internally generated selection: differences in distractor analysis and inhibition
Journal of Experimental Psychology: Human Perception and Performance
(1997) - et al.
Problem solving by patients with multiple sclerosis: comparison of performance on the Wisconsin and California Card Sorting Tests
Journal of the International Neuropsychological Society
(1996) - et al.
Attention in multiple sclerosis: correlates of impairment on the WAIS-R Digit Span Test
Applied Neuropsychology
(1995) - et al.
Evidence of reduced “cognitive inhibition” in schizophrenia
British Journal of Psychology
(1989) - et al.
Internal versus external cues and the control of attention in Parkinson’s disease
Brain
(1988)