Cognitive impairment in patients with Multiple Sclerosis (MS) has been reported across a range of cognitive domains (Rao et al., Reference Rao, Leo, Bernardin and Unverzagt1991). A host of studies have suggested that slowed information processing speed (IPS), i.e. an impaired ability to process information as quickly as healthy individuals, represents the key deficit underlying cognitive dysfunction in MS patients (DeLuca et al., Reference DeLuca, Chelune, Tulsky, Lengenfelder and Chiaravalloti2004).
Contrary to the idea that impaired IPS is a general or global deficit, evidences from different clinical populations have suggested that IPS may be a non-unitary construct (Chiaravalloti et al., Reference Chiaravalloti, Christodoulou, Demaree and Deluca2003; DeLuca & Kalmar, Reference DeLuca and Kalmar2007; Hsieh et al., Reference Hsieh, Chen, Wang and Lai2008; Shum et al., Reference Shum, Mcfarland and Bain1994). In a recent review about IPS in MS, it has been suggested that several steps compose information processing, starting with the input of information into the sensorial system and extending to the output (action or behavior). In this vein, Costa et al. (Reference Costa, Genova, DeLuca and Chiaravalloti2016) suggest that at least three major steps or processing components should be considered in order to disentangle the nature of IPS deficits in MS: (a) The transmission of sensorial information, (b) the completion of cognitive tasks, and/or (c) creating motor output. However, existing data are fragmented, including a limited number of components, which makes it difficult to achieve an integrative view on the nature of slowed information processing in MS. On the one hand, most investigations that compare performance of MS patients and healthy controls in simple visual detection Reaction Time (RT) tasks, have suggested slowness affecting the perceptual-motor component (De Sonneville et al., Reference De Sonneville, Boringa, Reuling, Lazeron, Adèr and Polman2002; Reicker et al., Reference Reicker, Tombaugh, Walker and Freedman2007). On the other hand, different authors found an inordinate increase of RT when increasing task complexity. This has been generally interpreted as an evidence of cognitive slowing independent of simpler perceptual and motor factors (Kujala et al., Reference Kujala, Portin, Revonsuo and Ruutiainen1994). In order to study specific cognitive mechanisms contributing to the so called cognitive component (e.g., attentional, memory, response selection, decision making, among others), different methods have been used such as subtraction (Parmenter et al., Reference Parmenter, Shucard and Shucard2007), percent change (Reicker et al., Reference Reicker, Tombaugh, Walker and Freedman2007; Tombaugh et al., Reference Tombaugh, Berrigan, Walker and Freedman2010), and covariance analyses (Laatu et al., Reference Laatu, Revonsuo, Hämäläinen, Ojanen and Ruutiainen2001). Results from these studies have led to describe a specific pattern of cognitive deficits in MS patients. For instance, focused attention (De Sonneville et al., Reference De Sonneville, Boringa, Reuling, Lazeron, Adèr and Polman2002), sustained attention (De Sonneville et al., Reference De Sonneville, Boringa, Reuling, Lazeron, Adèr and Polman2002; Stoquart-Elsankari et al., Reference Stoquart-Elsankari, Bottin, Roussel-Pieronne and Godefroy2010), and working memory (Parmenter et al., Reference Parmenter, Shucard and Shucard2007) seem to be slowed down in MS patients, while response selection (Archibald & Fisk, Reference Archibald and Fisk2000), interference control (Macniven et al., Reference Macniven, Davis, Ho, Bradshaw, Szabadi and Constantinescu2008), and divided attention (Stoquart-Elsankari et al., Reference Stoquart-Elsankari, Bottin, Roussel-Pieronne and Godefroy2010) appear to remain relatively unaffected. Unfortunately, other well-known processing components such as visual search (Neisser, Reference Neisser1964) or decision-making (Robertson et al., Reference Robertson, Manly, Andrade, Baddeley and Yiend1997; Whyte et al., Reference Whyte, Grieb-Neff, Gantz and Polansky2006) which may primarily modulate performance in many of the formerly mentioned cognitive tasks, have been scarcely explored in this population (Utz et al., Reference Utz, Hankeln, Jung, Lämmer, Waschbisch, Lee, Linker and Schenk2013). Moreover, no previous work has aimed to control for the influence of several different components in the same study by means of a comprehensive well-structured set of RT tasks, in an homogenous sample of MS patients.
Another important aspect regarding current investigations in MS concerns the role of clinical variables such as depression as a contributing factor to slowness of information processing. Clarifying this is of particular relevance given that depression is one of the most frequent psychiatric diagnoses in MS patients, affecting between 27% and 54% of all patients (Feinstein et al., Reference Feinstein, Magalhaes, Richard, Audet and Moore2014), and is known to impact IPS in non-MS depressed patients (Hammar, Reference Hammar2003). Different studies comparing performance of MS patients with and without depression have suggested that depression may exacerbate the slowness caused by MS (Arnett et al., Reference Arnett, Higginson, Voss, Wright, Bender, Wurst and Tippin1999; Lubrini et al., Reference Lubrini, Periáñez, Ríos-Lago and Frank2012; Lubrini et al., Reference Lubrini, Ríos Lago, Periañez, Tallón Barranco, De Dios, Fernández-Fourier, Díez Tejedor and Frank García2016). For instance, in an experimental design including four groups of MS patients with and without depressive symptoms (DS), depressed patients without MS, and healthy controls, Lubrini et al. (Reference Lubrini, Ríos Lago, Periañez, Tallón Barranco, De Dios, Fernández-Fourier, Díez Tejedor and Frank García2016) described a compounding effect where having MS and DS did worsen performance in the more demanding tasks, as compared to both having MS or depression alone. However, investigations trying to determine which processing components are slowed down in MS, have not considered the contributing role of depression to MS patients’ slowness. In this regard, it could be that depression may exacerbate the impairment of the components affected by MS or generate impairment in new ones. At this point, most prior studies have attempted to control for the effect of depression either by using samples with low or no depression, or by controlling its effect statistically, without disentangling the relative effect of MS and depression (Archibald & Fisk, Reference Archibald and Fisk2000; Parmenter et al., Reference Parmenter, Shucard and Shucard2007; Reicker et al., Reference Reicker, Tombaugh, Walker and Freedman2007; Tombaugh et al., Reference Tombaugh, Berrigan, Walker and Freedman2010).
Given the already demonstrated effect of MS in a set of RTs tasks (Lubrini et al., Reference Lubrini, Ríos Lago, Periañez, Tallón Barranco, De Dios, Fernández-Fourier, Díez Tejedor and Frank García2016), the first aim of this study was to identify which information processing components may underlie RT slowness in MS patients. Second, and given that DS seems to exacerbate cognitive slowness (Lubrini et al., Reference Lubrini, Ríos Lago, Periañez, Tallón Barranco, De Dios, Fernández-Fourier, Díez Tejedor and Frank García2016), we aimed to identify the components responsible for the additional RTs slowness associated with DS. In consonance with preceding literature, we first hypothesized that if RT slowness is not a generalized phenomenon, then differences between healthy controls and MS without DS will selectively affect some components with others being preserved. Second, we hypothesized that, if the compounding effect (i.e., the increasing RTs associated to depression in MS) is due to the worsening of those processing components already affected by MS, then differences between MS with and without DS will appear in the same components affected by MS. Alternatively, if the compounding effect is due to the impairment of new components, then differences between MS with and without DS will appear in those not affected by MS.
Method
Participants
Sixty-six right-handed MS patients were divided in two groups according to the presence of DS defined by a cut-off score ≥ 13 in the Spanish version of the Beck Depression Inventory (BDI; Sanz & Vázquez, Reference Sanz and Vázquez1998; Stoquart-Elsankari et al., Reference Stoquart-Elsankari, Bottin, Roussel-Pieronne and Godefroy2010). As a result, 33 MS patients were included in the group without DS and 33 patients in the group with DS. Given the questioned utility of the BDI in MS patients due to the high load of vegetative symptoms, this classification was tested with the method described by Strober and Arnett (Reference Strober and Arnett2010) was applied. Applying this methodology did not change the original classification. Also, 26 right-handed healthy controls (BDI score < 13) matched in age and education with the two groups of patients were included. Eligibility criteria for all participants were absence of severe motor or visual impairment that might interfere with testing; no premorbid history of learning disabilities; and no history of alcohol or drug abuse or nervous system disorder. Eligibility criteria for patients were: diagnosis of MS according to McDonald’s criteria (Polman et al., Reference Polman, Reingold, Edan, Filippi, Hartung, Kappos, Lublin, Metz, McFarland, O'Connor, Sandberg‐Wollheim, Thompson, Weinshenker and Wolinsky2005); diagnosis of a Relapsing Remitting course of MS; score on the Expanded Disability Status Scale (EDSS) < 6 (Kurtzke, Reference Kurtzke1983); and not having experienced a relapse or been treated with corticoids in the one-month time window before the experimental session. Demographic and clinical features of all participants are shown in Table 1. The Ethics Committee of the institution approved the study. Subjects were informed about the purpose of the investigation before the experimental session and signed a consent form according to the Declaration of Helsinki. The sample presented here partially overlaps the group studied in Lubrini et al. (Reference Lubrini, Ríos Lago, Periañez, Tallón Barranco, De Dios, Fernández-Fourier, Díez Tejedor and Frank García2016).
Table 1. Differences in Demographic and Clinical Variables between Participants
Note. BDI = Beck Depression Inventory; EDSS = Expanded Disability Status Scale; DS = depressive symptoms; SD = standard deviation; p = p value.
aComparison between MS without DS and healthy controls.b Comparison between MS with DS and MS without DS.
Experimental Tasks and Procedures
Participants were examined with five RT tasks (Fig. 1) within the context of a larger neuropsychological assessment. Evaluation was completed in a single session lasting between 70 and 90 min. All testing was performed using a PC with a 17-inches monitor that was controlled by Presentation® softwareFootnote 1. The order of task presentation was counterbalanced across participants. Both speed (RTs from correct trials) and accuracy (percent of correct responses) were measured in all tasks.
Figure 1. The figure illustrates from left to right a schematic example of stimuli sequence in the Simple Reaction Time (SRT), Simple Reaction Time-Sustained Attention to Response Task (SRT-SART), Choice Reaction Time (CRT), and Choice Reaction Time-Search (CRT-Search) tasks. Inter-Stimulus Interval (ISI) and stimulus duration (Stim. duration) in each task are specified.
Finger Tapping task (FT)
FT was used as a measure of motor function (Reitan & Wolfson, Reference Reitan and Wolfson1996). It has shown sensitivity to generalized slowing of responses (Strauss et al., Reference Strauss, Sherman and Spreen2006), involving motor performance with only minimal amounts of central resources required in more complex measures of speeded perceptual processing (Kennedy et al., Reference Kennedy, Clement and Curtiss2003). Reliability coefficient ranges from .58 to .93 in both healthy and clinical samples (Strauss et al., Reference Strauss, Sherman and Spreen2006). In MS studies, it has been explicitly used as a measure of motor speed (Stoquart-Elsankari et al., Reference Stoquart-Elsankari, Bottin, Roussel-Pieronne and Godefroy2010).
In the present experiment, following application norms described by Strauss et al. (Reference Strauss, Sherman and Spreen2006), participants were instructed to press repeatedly, and as fast as possible, the spacebar of a computer keyboard. The task consisted of five trials of 10-second duration each to be performed with the index finger of the dominant hand. The average time between two consecutive taps in the five trials was the dependent variable.
Simple Reaction Time (SRT)
SRT task was used as a measure of simple perceptual, motor, and sustained alertness processes (Jensen, Reference Jensen2006). According to this author, a larger fraction of the subjects’ SRT consists of the time needed for the sensory transduction and neural transmission of the stimulus to the brain and thence for the neural transmission to the muscles. In the MS literature, the SRT task has been used as one of the simplest and purest ways to assess IPS. Consequently, many studies use it as a baseline or control measure of more complex processing speed tasks (Reicker et al., Reference Reicker, Tombaugh, Walker and Freedman2007; Stoquart-Elsankari et al., Reference Stoquart-Elsankari, Bottin, Roussel-Pieronne and Godefroy2010; Tombaugh et al., Reference Tombaugh, Berrigan, Walker and Freedman2010). SRT reliability tends to be high given that it shows virtually no practice effect after the first 10 trials (Teichner & Krebs, Reference Teichner and Krebs1974). In the present experiment, following application norms described by Reicker et al., (Reference Reicker, Tombaugh, Walker and Freedman2007), participants were instructed to press the left button of the mouse with the index finger of their dominant hand as quickly as possible when the stimulus “+” appeared in the center of the computer screen. The task consisted of 50 trials with a total task duration of 2–3 min.
Simple Reaction Time-Sustained Attention to Response Task (SRT-SART)
SRT –SART task was used as a measure of response strategy-inhibition (Carter et al., Reference Carter, Russell and Helton2013). Significant Pearson’s product-movement correlations have been found between SART RT to correct trials, and No Go errors (r = –.82; p < .001; Seli, Reference Seli2016), suggesting that both scores provide useful measures of response strategy-inhibition. In the present experiment, following application norms described by Robertson et al. (Reference Robertson, Manly, Andrade, Baddeley and Yiend1997), participants were instructed to press the left button of the mouse with the right index finger whenever a stimulus (a digit between 1 and 9) appeared in the center of the screen, but not to press if the digit shown was number 3. The task consisted of 189 trials (21 of them were No Go trials) with a total task duration of approximately 4 min.
Choice Reaction Time (CRT)
CRT task was used as a measure of visual perceptual decision time. It has been related to the same processes involved in the SRT plus the processing of uncertainty as to which one of the stimulus would appear next, that is, decisional processing (Jensen, Reference Jensen2006; Pipingas et al., Reference Pipingas, Harris, Tournier, King, Kras and Stough2010). Test-retest reliability coefficient scored .81 (Pipingas et al., Reference Pipingas, Harris, Tournier, King, Kras and Stough2010). The CRT task has been used in previous research on IPS in MS (Reicker et al., Reference Reicker, Tombaugh, Walker and Freedman2007; Stoquart-Elsankari et al., Reference Stoquart-Elsankari, Bottin, Roussel-Pieronne and Godefroy2010; Tombaugh et al., Reference Tombaugh, Berrigan, Walker and Freedman2010). In this task, derived from Visual Choice Reaction Time task (Chiaravalloti et al., Reference Chiaravalloti, Christodoulou, Demaree and Deluca2003), participants were instructed to press the left or the right button of the mouse every time a square or a circle, respectively, appeared in the middle of the screen. The task consisted of 80 trials. Total task duration was approximately 3 min.
Choice Reaction Time-Search (CRT-Search)
CRT-Search task was used as a measure of visual search (Neisser, Reference Neisser1964). In this task, derived from Neisser’s paradigms (Reference Neisser1964), participants were told to press the left or right button of the mouse depending respectively on the presence or absence of the letter “Z” in a string of six letters. The task consisted of 128 trials. Total task duration varied between 5 and 8 min. Stimuli were classified according to two different dimensions: Presence/absence of the letter “Z” (Type of Stimulus: Target vs. Non Target), and the visual features of the letters in the string (rounded or angular; Level of Interference: Low vs. High, respectively). Thus, the combination of Type of Stimulus and Level of Interference resulted in four different trial types: Target-Low Interference (e.g., GODZCQ); Target-High Interference (e.g., VWMZEX); Non Target-Low Interference (e.g., CQUGRD); Non Target-High Interference (e.g., VXWEIM). A visual search component can be measured by comparing RTs in the “Non Target-Low Interference” condition of the CRT-Search task with RTs in the “Target-Low Interference” condition. At this regard, “Target-Low Interference” and “Non-Target-Low Interference” conditions differ in the difficulty of the search, as described in preceding experimental research (Neisser, Reference Neisser1964; Treisman, Reference Treisman1988). Accordingly, the “Target-Low Interference” condition would involve a fast, parallel, and unlimited capacity pre-attentive processing since the presence of the target flanked by dissimilar distracters will allow a “pop-out” detection of its basic features. On the contrary, the lack of targets in the “Non-Target-Low Interference” condition would impel a subsequent slower, serial spatially-limited attention-demanding processing, thus increasing RT (Treisman, Reference Treisman1988).
Analysis
Group differences in demographic and clinical variables (sex, age, education, estimated premorbid IQ (Bilbao & Seisdedos-Cubero, Reference Bilbao and Seisdedos-Cubero2004), disease duration, EDSS, and BDI score) were determined by means of t-tests or Chi square statistical tests where appropriate. The total number of medications with a psychomotor effect was also compared between the two groups of patients. Drug categories considered for this analysis included antispasmodics and muscle relaxants, tricyclic antidepressants, anticholinergics, and benzodiazepines (Arnett et al., Reference Arnett, Higginson, Voss, Wright, Bender, Wurst and Tippin1999). Pearson’s product-moment correlations analyses between mean RT and percentage of correct responses per task for each group were performed as a precaution against speed-accuracy trade off phenomenon accounting for RT modulations (De Sonneville et al., Reference De Sonneville, Boringa, Reuling, Lazeron, Adèr and Polman2002).
In order to explore the first hypothesis regarding the processing components accounting for RT slowness in MS, comparisons were performed between healthy controls and MS patients without DS. In order to explore the second hypothesis regarding the processing components being affected by the compounding effect, comparisons were performed between MS patients with and without DS. Between group comparisons in the different components were performed by means of a series of independent sample t-tests and ANCOVAs according to the following rationale:
1. The presence of information processing slowness associated to a “motor” component was analyzed by means of independent sample t-tests comparing the FT task between groups.
2. The presence of information processing slowness associated to a perceptual, motor, and sustained alertness components, henceforth referred to as “perceptual-motor-alertness”, were analyzed by means of independent sample t-tests comparing RT to the SRT task between groups.
3. The presence of information processing slowness associated to a “response strategy-inhibition” component was analyzed by means of an ANCOVA with RT in the SRT-SART task as the dependent variable and RT in the SRT task as the covariate. Using SRT as a covariate allowed controlling for perceptual, motor, and sustained alertness components shared with the SRT-SART task.
4. The presence of information processing slowness associated to a visual perceptual decision time component henceforth referred to as “decisional”, was analyzed by means of an ANCOVA with RTs in the CRT task as the dependent variable and RT in the SRT task as the covariate. At this regard, CRT has been related to the same processes involved in the SRT plus the decisional processing (Jensen, Reference Jensen2006; Pipingas et al., Reference Pipingas, Harris, Tournier, King, Kras and Stough2010).
5. The presence of information processing slowness associated to a “visual search” component was analyzed by means of an ANCOVA with RTs in the “Non Target-Low Interference” condition of the CRT-Search task as the dependent variable and RTs in the “Target-Low Interference” condition as the covariate. Using the “Target-Low Interference” condition as the covariate allowed controlling for common perceptual, motor, and cognitive processes except for serial visual search as measured in the “Non Target-Low Interference” condition.
A significance level of p < .05 was adopted for all analyses. All analyses were performed using SPSS v22.0.
Results
Sample Characteristics and Preliminary Analyses
Table 1 describes participants’ demographic and clinical characteristics. No differences were found between MS patients without DS and healthy controls in sex, χ2(1) = .001, p = .969; age, t(57) = 0.4, p = .666; education, t(57) = –1.8, p = .07; estimated premorbid IQ, t(57) = –.9, p = .368, or BDI score, t(57) = 1.3, p = .212. The two groups of MS patients did not differ significantly in any demographic or clinical variable, p > .157 in all cases, except for BDI score, p < .001. Accuracy was high across tasks and groups ranging between 88% and 99.3% of correct responses (Table 2). Results of correlation analyses between accuracy and RTs revealed a lack of significant correlations in the MS patients without DS (p > .095 in all cases), MS patients with DS (p > .406 in all cases), and healthy controls (p > .054) with the only exception of the SRT task in the latter group (r = .539; p = .005).
Table 2. Means (Standard Deviation) of Reaction Times (RT) in Milliseconds and Percent of Correct Responses (% correct) for the RT Tasks in the Three Groups: MS Patients with Depressive Symptoms, MS Patients without Depressive symptoms, and Healthy Controls
Note. DS = depressive symptoms; FT = Finger Tapping; SRT = Simple Reaction Time task; SRT-SART = Simple Reaction Time-SART task; CRT = Choice Reaction Time task; CRT-Search = Choice Reaction Time-Search task; Int = Interference.
a Conditions of the CRT-Search task.
Reaction Times (RTs)
Comparisons between MS patients without DS and healthy controls revealed that, in the FT task, MS patients exhibited increased response times as compared to healthy controls, t(57) = 2.9, p = .006, d = .77, confidence interval 95% CI [5.08, 28.6]. Between group comparison in the SRT task revealed that MS patients without DS had increased RT with respect to healthy controls, t(57) = 2.1, p = .036, d =.57, confidence interval 95% CI [2.1, 60.27]). The ANCOVA designed to measure the visual search component (as measured in Non Target-Low Interference condition of the CRT-Search task using Target-Low Interference condition as the covariate) revealed that MS patients had slower responses as compared to healthy controls, F(1, 56) = 4.5, p = .039, d = .28, 95 95% CI [3.33, 128.14]. On the other hand, there were no differences between these groups in the response strategy-inhibition component as measured by the SRT-SART task using the SRT task as covariate, F(1, 56) = 0.9, p = .342. d = .13, 95% CI [–11.41, 32.3], nor in the decisional component as measured by the CRT task using the SRT task as covariate, F(1, 56) = 0.4, p = .508, d = .09, 95% CI [–17.37, 34.67], see Table 2, and Figure 2).
Figure 2. Comparisons between Multiple Sclerosis Patients without Depressive Symptoms (MS without DS), and Healthy Controls in Different Components of the Stimulus-Response Pathway (Motor, Perceptual-Motor-Alertness, Response Strategy-Inhibition, Decisional, and Visual Search). Asterisks indicate statistically significant differences between groups; Mot = motor; Perc-Mot-Alert = perceptual-motor-alertness; Resp Strat-Inhib = response strategy-inhibition.
Comparisons between MS patients with and without DS revealed that, in the FT task, MS patients with DS exhibited increased response times as compared to MS patients without DS, t(52) = –3.1, p = .003, d = .77, confidence interval 95% CI [–43.6, –9.6]. Between group comparison in the SRT task revealed that there were no differences between the groups, t(64) = –.6, p = .53, d = .16, 95% CI [–49.51, 25.71]. The ANCOVA designed to measure the decisional component (as measured in the CRT task using SRT as the covariate) revealed that MS patients with DS had slower responses as compared to MS patients without DS, F(1, 63) = 5.1, p = .027, d = .28, 95% CI [–93.73, –5.86]. No differences were found between groups in the response strategy-inhibition, F(1, 66) = 0.04, p = .849, d =.03, 95% CI [–24.98, 20.61], nor in the visual search components, F(1, 63) = 1.8, p = .184, d = .17, 95% CI [–122.97, 24.14] (see Table 2, and Figure 3).
Figure 3. Comparisons between Multiple Sclerosis Patients with and without Depressive Symptoms (MS with, and without DS) in Different Components of the Stimulus-Response Pathway (Motor, Perceptual-Motor-Alertness, Response Strategy-Inhibition, Decisional, and Visual Search). Asterisks indicate statistically significant differences between groups. Mot = motor; Perc-Mot-Alert = perceptual-motor-alertness; Resp Strat-Inhib = response strategy-inhibition.
Discussion
Slowness of information processing has been well established as a primary deficit in MS (DeLuca et al., Reference DeLuca, Chelune, Tulsky, Lengenfelder and Chiaravalloti2004; Lubrini et al., Reference Lubrini, Ríos Lago, Periañez, Tallón Barranco, De Dios, Fernández-Fourier, Díez Tejedor and Frank García2016). The aim of this study was to identify which information components of the stimulus-response processing pathway may underlie RT slowness in MS patients with and without DS. Following preceding three-factorial proposals (Costa et al., Reference Costa, Genova, DeLuca and Chiaravalloti2016), motor, perceptual, and cognitive components were studied as potential sources of slowness. Innovatively, this was done by means of a comprehensive set of independent sample t-tests, and analyses of covariance using five RT paradigms. One of the advantages of using ANCOVA as an analytical strategy is that it does not assume any specific functional architecture about the cognitive system i.e., either serial or parallel. Beside alternative methods being used to control the influence between cognitive components (for example, subtraction or percent change methods), the ANCOVA estimates and controls for the amount of shared variance observed between the dependent variable and the covariate.
Three of the five explored components were found to be slowed down in MS patients as compared to controls. On the one hand, both motor and perceptual-motor-alertness components, as measured in a FT and a SRT tasks, respectively, exhibited a specific impairment in the MS group. These results are consistent with those suggesting that perceptual-motor factors may account for slowness in SRT tasks (Reicker et al., Reference Reicker, Tombaugh, Walker and Freedman2007). On the other hand, the visual search cognitive component also revealed differences between MS patients and controls. Thus, MS seems to affect the ability to orient attention for establishing and retaining an efficient search strategy. At this regard, visual search paradigms have not been frequently used in MS. However, the present results agree with those from RT studies demonstrating impairment on the visual search component in MS using a featured-based visual search paradigm (Utz et al., Reference Utz, Hankeln, Jung, Lämmer, Waschbisch, Lee, Linker and Schenk2013), and are in consonance with prior neuropsychological evidence derived from attention tests involving visual search demands, such as the Trail Making Test (Leavitt et al., Reference Leavitt, Wylie, Krch, Chiaravalloti, DeLuca and Sumowski2014), and the Symbol Digit Modalities Test (Arnett et al., Reference Arnett, Higginson, Voss, Wright, Bender, Wurst and Tippin1999). Response strategy-inhibition, and decisional components resulted to be relatively preserved in MS patients, in line with earlier reports (Macniven et al., Reference Macniven, Davis, Ho, Bradshaw, Szabadi and Constantinescu2008; Reicker et al., Reference Reicker, Tombaugh, Walker and Freedman2007). Taken together, the results showed that different processing components were selectively affected by MS, thus confirming the first hypothesis about the non-generalized nature of slowness of information processing.
In order to clarify whether the compounding effect is due to the worsening of processing components already affected by MS or to the impairment of new components, as established by the second hypothesis, patients with and without DS were compared by means of the same set of t-tests, and ANCOVAs described previously. Two of the five explored components revealed to be impaired in MS patients with DS. While DS did not generate differences in the perceptual-motor-alertness component, differences were found when the motor component was analyzed alone. This result is in consonance with data from non-MS depressed populations showing that depression deteriorates motor speed (Caligiuri & Ellwanger, Reference Caligiuri and Ellwanger2000). Second, while the response strategy-inhibition component was preserved, a marked decline in performance was found in the decisional component. This result indicates that patients with DS were able to mobilize resources to complete controlled tasks so long as a decision between different responses was not required. However, when task demands involved to select between alternative responses (i.e., CRT task), a marked decline in performance was observed (Thomas et al., Reference Thomas, Goudemand and Rousseaux1999). None of the remaining components resulted to be affected in the MS with DS group as compared to MS without DS group. Lastly, correlation analyses between mean RTs and the percentage of correct responses in each task for each group revealed, in general, an absence of significant positive correlations that allowed to reject the possibility of an “speed accuracy trade off” effect accounting for the described RT effects (De Sonneville et al., Reference De Sonneville, Boringa, Reuling, Lazeron, Adèr and Polman2002). Taken together, the results showed that the compounding effect is in part due to the worsening of the motor component already affected by MS, and in part due to impairment of the decisional component, which was not primarily affected by MS.
In summary, and contrary to the idea that slowed IPS is a general/global deficit in MS, the present findings suggest that dissociable components of the stimulus-response pathway differentially contributed to patients’ slowed performance. While perceptual-motor-sustained alertness (as measures by the FT and STR tasks), and visual search components (as measured by CRT-Search) seemed to be slowed down by MS, the compounding effect of MS and DS exacerbated deficits in the motor component, and slowed down the decisional component. These results have relevant implications for both theoretical models and clinical practice. First, they could help in designing assessment and treatments oriented to specific processing stages according to patients’ cognitive profiles. Second, the results suggest that, at least in early stages of MS, targeting mood might represent a potential way to address IPS deficits. This is quite promising given that there is evidence suggesting the effectiveness of both psychological and pharmacological approaches for depression in MS (Feinstein et al., Reference Feinstein, Magalhaes, Richard, Audet and Moore2014). Moreover, understanding the nature of cognitive dysfunction in MS would be incomplete if the interplay between IPS and DS is not considered. Lastly, taking into account the different subcomponents of the information processing pathway, it seems crucial to describe the causal mechanisms of slowed IPS in MS. As limitations, even when the present report included a wide range of processing components and tasks, further cognitive components using a wider range of tasks as well as the possible interactions between them should be studied in the future. Another limitation, and considering the small sample size, non-significant differences founded should be addressed in the future with larger samples in order to increase the power of analyses.
Lastly, and given the potential influence of anxiety on certain IPS measures as evidences by recent researches (Goretti et al., Reference Goretti, Viterbo, Portaccio, Niccolai, Hakiki, Piscolla, Iaffaldano, Trojano and Amato2014; Morrow, Rosehart, & Pantazopoulos, Reference Morrow, Rosehart and Pantazopoulos2016), this variable should be addressed in forthcoming studies on MS.
Conflicts of Interest:
Conflicts of Interest. None.
Funding Statement:
This research received no specific grant from any funding agency, commercial or not-for-profit sectors.
