Parkinson's disease (PD) has been associated with facial expression recognition difficulties. However, this impairment could be secondary to the one produced in other cognitive processes involved in recognition, such as selective attention. This study investigates the influence of two selective attention components (inhibition and visual search) on facial expression recognition in PD. We compared facial expression and non-emotional stimuli recognition abilities of 51 patients and 51 healthy controls, by means of an adapted Stroop task, and by “The Face in the Crowd” paradigm, which assess Inhibition and Visual Search abilities, respectively. Patients scored worse than controls in both tasks with facial expressions, but not with the other nonemotional stimuli, indicating specific emotional recognition impairment, not dependent on selective attention abilities. This should be taken into account in patients' neuropsychological assessment given the relevance of emotional facial expression for social communication in everyday settings.

Introduction

Parkinson's disease (PD) is a neurodegenerative, chronic and progressive condition associated with a loss of dopaminergic neurons in the substancia nigra (pars compacta) of the basal ganglia (Hammond, Bergman, & Brown, 2007). It is predominantly characterized by motor impairments such as tremor, bradykinesia, rigidity, and postural instability (Jankovic, 2008). In addition to these motor symptoms, patients often suffer from cognitive impairments, especially those relating to visuospatial perception, memory and executive functions (Alonso-Recio, Martín, Carvajal, Ruiz, & Serrano, 2013; Barone et al., 2011; Kehagia, Barker, & Robbins, 2010). Most of this cognitive symptomatology seems to be similar to that shown by individuals with frontal lobe damage (Lezak, Howieson, & Loring, 2004).

In addition to cognitive symptoms, emotional disturbances have also been described in patients with PD. These difficulties seem to be reflected in their global mood state, increasing the risk of anxiety or depressive disorders (for a review see Chaudhuri, Healy, & Schapira, 2006; Merello, 2008); but they may also involve more specific abilities to express and recognize emotions (Ariatti, Benuzzi, & Nichelli, 2008; Dara, Monetta, & Pell, 2008; Schröder et al., 2006; Simons, Ellgring, & Pasqualini, 2003). Thus, together with general difficulties in prosodia and emotional facial expression (EFE), more specific EFE recognition disabilities have been found in PD (Gray & Tickle-Degnen, 2010; Péron, Dondaine, LeJeune, Grandjean, & Verin, 2012). Facial expressions play a very relevant role in social interaction, as an important and very common way to communicate and recognize other people's intentions and desires (Adolphs, 2002). Hence, to establish the possible decline in these abilities in PD may be important to assess and, eventually, design specific intervention programs to prevent their communication difficulties.

Several controversial aspects to this EFE recognition disability have been identified, such as whether it concerns all the emotions or whether it is predominant in some, in particular the negative ones (Assogna et al., 2010; Beatty et al., 1989; Breitenstein, Daum, & Ackermann, 1998; Dujardin et al., 2004; Herrera, Cuetos, & Rodriguez-Ferreiro, 2011; Kan, Kawamura, Hasegawa, Mochizuki, & Nakamura, 2002; Lachenal-Chevallet et al., 2006; Suzuki, Hoshino, Shigemasu, & Kawamura, 2006; Yip, Lee, Ho, Tsang, & Li, 2003). Another controversial point refers to the possibility that EFE recognition impairments in PD are not genuine and independent deficits, but are secondary to other cognitive (i.e., executive) problems (for a review, see Alonso- Recio, Serrano- Rodriguez, Carvajal- Molina, Loeches- Alonso, & Martín- Plasencia, 2012; Assogna, Pontieri, Caltagirone, & Spalleta, 2008; Gray & Tickle-Degnen, 2010). It is, also, noteworthy that emotional recognition, as it is commonly assessed, involves not only identification of the affective message expressed by the face, but also the starting of executive processes (e.g., categorization, working memory, and/or attention). A possible impairment in these cognitive processes may therefore influence emotional recognition performance. This relationship between executive function and EFE recognition abilities in PD has been explored in some studies about categorization and working memory with conflicting results (Alonso-Recio, Martín, Rubio, & Serrano, 2013; Alonso-Recio, Martín-Plasencia, Loeches-Alonso, & Serrano-Rodríguez, 2014; Clark, Neargarder, & Cronin-Golomb, 2008; Cohen, Gagne, Hess, & Pourcher, 2010; Pell & Leonard, 2005).

Apart from categorization or working memory abilities, other aspects of executive function may influence EFE recognition, such as attention processes. In fact, these have been referred to as among the most frequent executive problems found in PD patients (Muslimovic, Post, Speelman, De Haan, & Schmand, 2009). More specifically, and following Sohlberg and Mateer's (1987) clinical framework on attention organization, PD patients show relatively preserved basic attention abilities, but they are impaired in task engaging divided and selective attention (Dujardin et al., 2013). Moreover, it has been shown that in selective attention PD patients may show deficits in tasks involving inhibition and visual search abilities, which may convey problems for resource allocation (Djamshidian, O'Sullivan, Lees, & Averbeck, 2011; Horowitz, Choi, Horvitz, Côté, & Mangels, 2006; Hsieh, Chen, Wang, & Lai, 2008; Mannan, Hodgson, Husain, & Kennard, 2008). These impairments seem to be related to the characteristic dopaminergic deficit of PD individuals. In support of this assumption, it has been observed that differences between patients and healthy individuals in attention tasks disappear when patients are under dopamine therapy and, also, after deep brain stimulation (Djamshidian et al., 2011; Fera et al., 2007; Jahanshahi et al., 2000).

The ability to recognize EFEs may depend, at least in part, on the ability to process different sources and kinds of information at the same time, and to attend to the relevant facial-affective features while inhibiting irrelevant ones (García-Rodríguez et al., 2012; Vuilleumier, 2002). To our knowledge, studies on the direct influence of attention in EFE recognition in PD have only been carried out by García-Rodríguez and colleagues (2012) and focused on divided attention. This study found that EFE recognition in PD patients was only compromised when emotional stimuli were processed concurrently with other interfering ones. With regards to selective attention, only Assogna and colleagues (2010) reported indirect evidence for this possible relationship from the significant correlations they found between EFE recognition abilities and the Stroop test scores for their PD group. However, it is noteworthy that this significant correlation provides only indirect support for the possible implication of one of the selective attention components (inhibition ability) in EFE recognition in PD. Furthermore, they found only a partial deficit in recognition of the facial expression of disgust.

Taking into account this very limited evidence, the objective of the present study was to contribute to our knowledge of the relevance of selective attention processes in EFE recognition abilities in PD patients. For this purpose, we designed two experimental settings which allowed us to compare two different components of selective attention: inhibition and visual search abilities. In each of them, we compared PD patients and healthy individuals' performances with EFEs and other nonemotional stimuli. The first setting, designed to measure inhibition abilities, was based on the Stroop effect paradigm (Stroop, 1935). Here, we compared both groups in a classic color-word Stroop task and in a novel EFE-word task. The second setting, designed to assess visual search abilities, was based on “a face-in-the-crowd effect” paradigm (Hansen & Hansen, 1988). In this case, we contrasted PD and healthy individuals' searching abilities with nonemotional faces and with EFEs. From these comparisons, we hoped to discern whether EFE recognition abilities in PD are related or not to selective attention components.

Method

Participants

Participants were comprised of 51 individuals (31 women) with idiopathic PD, diagnosed by neurologists on the basis of international guidelines (Hughes, Ben-Shlomo, Daniel, & Lees, 2001), and 51 healthy controls (HC; 29 women). Common exclusion criteria for PD patients and HC group were the presence of major neurological disorders, major psychiatric disorders, visual deficits, and suspected dementia or cognitive impairment (MMSE < 27; Folstein, Folstein, & Mchugh, 1975). Specific exclusion criteria for PD patients were the presence of an unclear history of chronic dopaminergic treatment responsiveness. Background data for the two groups are summarized in Table 1 where it can be observed that both groups did not significantly differ for sex, age, educational level, and general cognitive abilities, as estimated by the Spanish version of National Adult Reading Test (Test de Acentuación de Palabras, TAP; Del Ser, González-Montalvo, Martinez-Espinosa, Delgado-Villapalos, & Bermejo, 1997) and by the MMSE. However, groups showed a statistically significant difference with regard to the Spanish version of Geriatric Depression Scale, GDS-R (Izal, Montorio, Nuevo, Perez-Rojo, & Cabrera, 2010). However, neither PD patients nor HC had scores compatible with depression (GDS-R > 5).

Table 1.

Major characteristics of 51 HC and 51 participants with PD

Characteristics HC
 
PD
 
χ2 or t p 
 M SD M SD 
Age (years) 64.82 64.72 64.72 3.98 0.87 .93 
Gender (male): n (%) 22 (43) 20 (39) 0.16 .69 
Education     3.59 .61 
 No studies 7.9 5.9   
 Basic studies 23.5 13.7   
 Primary studies 37.3 43.1   
 Secondary studies 19.6 17.6   
 Higher studies 11.8 19.6   
TAP 22.29 2.61 22.12 5.95 0.15 .88 
MMSE 29.17 1.03 29.06 1.01 0.58 .56 
GDS-R 1.63 1.62 2.59 2.64 −2.21 <.05 
Characteristics HC
 
PD
 
χ2 or t p 
 M SD M SD 
Age (years) 64.82 64.72 64.72 3.98 0.87 .93 
Gender (male): n (%) 22 (43) 20 (39) 0.16 .69 
Education     3.59 .61 
 No studies 7.9 5.9   
 Basic studies 23.5 13.7   
 Primary studies 37.3 43.1   
 Secondary studies 19.6 17.6   
 Higher studies 11.8 19.6   
TAP 22.29 2.61 22.12 5.95 0.15 .88 
MMSE 29.17 1.03 29.06 1.01 0.58 .56 
GDS-R 1.63 1.62 2.59 2.64 −2.21 <.05 

Note: M = mean; SD = standard deviation, TAP = Test de acentuación de Palabras, MMSE = Mini-Mental State Examination; GDS-R = Geriatric Depression Scale.

Patients were recruited from three Parkinson Associations of Madrid (Spain). Median of PD severity on the Hoehn and Yahr Scale (Hoehn & Yahr, 1967) was 2 (range 1–4). On the Clinical Impression of Severity Index for Parkinson's Disease (CISI-PD; Martinez-Martin, Forjaz, Cubo, Frades, & de Pedro Cuesta, 2006), the PD group mean score was 9.12 out of a possible maximum of 24 (SD = 3.49). Global functional capacity and dependence corresponded to 83.72/100 (SD = 11.82) as measured by the Schwab and England scale (Schwab and England, 1969) and to 40.16/156 (SD = 29.05) on the Spanish version of Parkinson Disease Questionnaire (PDQ-39; Martinez-Martin & Frades Payo, 1998). The mean illness duration was 6.32 years (SD = 3.87). All patients were tested after being administered anti-Parkinsonian medication, distributed as follows: carbidopa/l-DOPA (45), D2 agonists (42), MAO inhibitors (26), amantadine (3), and anticolinergic agents (1). They were tested in the morning after medication had been administered (on-state).

Participants were informed of the confidential and anonymous treatment of their data and signed the informed consent. The study was completed in accordance with the Helsinki Declaration and approved by the Ethical Committee of the Universidad Autónoma de Madrid.

Experimental Tasks

Inhibition task

We created two set of stimuli following the “Stroop effect” paradigm, one with EFE-word and one with Color-word. For the EFE-word set, 40 pictures of males and females faces (20 men) showing happy, anger, fear, disgust, or sadness (eight faces for each emotion) were selected from the FACES Database (Ebner, Riediger, & Linderberger, 2010). The hair and background was removed from all pictures, in order to eliminate insignificant or distracting information for EFE recognition. We then composed a total of 40 trials in which EFEs appeared with a superimposed emotion category name. For the eight examples of each EFE, we generated four congruent stimuli (the emotion category name corresponded to the EFE shown) and four with an incongruent superimposed name (not corresponding to the EFE shown). In these incongruent trials, EFE and superimposed name were balanced to cover all the possible combinations (see Fig. 1, example A). Participants were instructed to identify the EFE and to ignore the superimposed emotion category name. For the Color-word set we created 40 pictures with the name of a color (eight examples of each of RED, GREEN, BLUE, YELLOW, or WHITE colors) written in capital letters and filled with a color ink. For the eight examples of each of the five names of colors, we generated four congruent examples (filled with the same color ink) and four incongruent ones (filled with a different color ink). As in the EFE-word task, the color words and the color ink were balanced (see Fig. 1, example B). Participants were instructed to identify the ink color, and to ignore the color name.

Fig. 1.

Examples of stimuli used in Inhibition task. (A) Facial expression. (B) Color.

Fig. 1.

Examples of stimuli used in Inhibition task. (A) Facial expression. (B) Color.

Visual search task

We designed two set of stimuli (EFEs and nonemotional faces) following the “a face-in-the-crowd effect” paradigm. For the EFE set (EFEs crowd), 30 pictures of 15 different male and 15 female faces displaying happiness, anger, fear, disgust, and sadness (six examples of each EFE) were selected from the FACES Database (Ebner et al., 2010). We composed of 20 items with these stimuli. Each of them consisted in a matrix with 6 (horizontal) × 4 (vertical) EFEs, then showing a total of 24 examples of EFEs. Half of the trials comprised 24 copies of the same example (target-absent trials), and the remaining were composed of 23 copies of the same EFE plus another one showing a different EFE (target trials) (Example A on Fig. 2). Participants were instructed to decide whether all the 24 faces showed the same EFE or whether any displayed a different one. Each EFE appeared six times, two in target-absent trials and four in target trials (two as target and two as nontarget). For the nonemotional faces set (nonemotional faces crowd), 30 pictures of men (15) and women (15) displaying neutral expression were selected from the FACES Database (Ebner et al., 2010). With these pictures we composed 20 items consisting in a matrix with 6 × 4 nonemotional faces (10 target-absent and 10 target trials) (Example B in Fig. 2). Participants had to decide whether the 24 nonemotional faces corresponded to the same person or to a different person.

Fig. 2.

Examples of stimuli used in Visual Search task. (A) Facial identity. (B) Facial expression.

Fig. 2.

Examples of stimuli used in Visual Search task. (A) Facial identity. (B) Facial expression.

Procedure

All experimental tasks were created with E-prime 1.2 (Schneider, Eschman, & Zuccolotto, 2002) which controlled stimuli presentation and trial randomization. Participants were individually tested in a quiet room by means of a high-resolution computer monitor at a visual distance of 60 cm. The PD group was assessed at a time of day when their motor symptoms were less severe (“on-state”).

The tasks were completed in a fixed order in two different sessions of 1 hr duration each, and with an interval of 2 weeks between them (Inhibition task in the first session and Visual Search task in the second) to avoid participants' fatigue. Emotional face and other trials were blocked within a session. Two PD patients and two individuals from the HC group did not attend the second session, so they did not complete the second task. In both tasks, participants were instructed to respond as soon as possible by pressing the corresponding number to the selected answer on the computer keypad. Once the response had been made (or 10,000 ms elapsed), a fixation point appeared again and a new trial started. The order of the trials in each task was randomized among participants.

Each task started with a verbal explanation and included several practice trials which can be repeated until the task was fully understood. The experiments began showing a centrally located white cross in the middle of the screen for 500 ms, to fixate the participant's attention. After this, the first trial was shown and remained until a response was made (or 10,000 ms elapsed). For the Inhibition task, the five response options appeared in the right of the screen (see Fig. 1) whereas for Visual Search task, the two response options were shown in the bottom of the screen (see Fig. 2).

Results

Inhibition Task

Accuracy

For the Inhibition task, accuracy mean scores (see Table 2) were analyzed by a two-way mixed ANOVA, with Group (PD patients vs. HC) as the between-subjects factor and Stimuli (EFE-word and Color-word) as the within-subjects factor. The analysis revealed a significant main effect of Stimuli (F(1,100) = 255.26, p < .001, η2 = .72), showing that all participants were more accurate in Color-word than in EFE-word; and Group (F(1,100) = 6.84, p < .05, η2 = .06), with PD patients being less accurate than HC in all sets. The analysis also showed that there was a significant interaction between Group and Stimuli (F(1,100) = 5.84, p < .05, η2 = .06). Analysis for simple effects showed a statistically significant difference when comparing PD and HC scores for EFE-word (F(1,100) = 7.36, p < .01, η2 = .07), but not for Color-word (F(1,100) = 0.25, p = .62, η2 = .002). These results indicated that PD patients were less accurate in their responses than the HC group in EFE-word (see Table 2).

Table 2.

Mean and standard deviations for accuracy scores, reaction times, and omitted responses in the inhibition task, both for EFE-word and color-word sets, by HC and PD patients

  Set HC (N = 51)
 
PD (N = 51)
 
Average 
M SD M SD 
Accuracy (/40) EFE-word 32.69 3.61 30.29 5.16 31.49 
Color-word 38.90 1.50 38.73 2.03 38.81 
Average 35.79  34.51   
Reaction times (/10,000) EFE-word 5,761.66 626.46 5,776.39 793.53 5,769.03 
Color-word 5,807.82 663.84 5,831.00 741.45 5,819.41 
Average 5,784.74  5,803.70   
Omitted responses (/40) EFE-word 1.04 1.13 1.71 2.05 1.37 
Color-word 0.24 0.55 0.19 0.49 0.22 
Average 0.64  0.95   
  Set HC (N = 51)
 
PD (N = 51)
 
Average 
M SD M SD 
Accuracy (/40) EFE-word 32.69 3.61 30.29 5.16 31.49 
Color-word 38.90 1.50 38.73 2.03 38.81 
Average 35.79  34.51   
Reaction times (/10,000) EFE-word 5,761.66 626.46 5,776.39 793.53 5,769.03 
Color-word 5,807.82 663.84 5,831.00 741.45 5,819.41 
Average 5,784.74  5,803.70   
Omitted responses (/40) EFE-word 1.04 1.13 1.71 2.05 1.37 
Color-word 0.24 0.55 0.19 0.49 0.22 
Average 0.64  0.95   

Given the differences we found between PD and HC in depressive mood (measured by the GDS-R scale), we performed an additional covariance analysis in an attempt to discern whether this could explain the differences in accuracy to EFEs. Analysis revealed that GDS-R (included as a covariate) was not significant (F(1,99) = 0.21, p = .65, η2 = .002) and that the significant interaction Group × Stimuli was maintained with respect to the previous one (F(1,99) = 5.92, p < .05, η2 = .06), indicating that differences between PD and HC in response to EFE-word were not influenced by their differences in depression scores.

Finally, in order to analyze possible different effects in the accuracy scores for any of the EFE (see Table 3) we performed a mixed ANOVA with Group (PD patients vs. HC) as the between-subjects factor and emotion (happiness, sadness, anger, disgust, and fear) as the within-subjects factor. The analysis revealed that interaction was not statistically significant (F(4,400) = 3.82, p = .15, η2 = .02) indicating that differences between PD and HC were not attributable to specific deficits with any of the emotions we observed.

Table 3.

Mean and standard deviation for accuracy scores in each emotion in EFE-word set, by HC and PD patients

 HC
 
PD
 
 M SD M SD 
Happiness 7.90 0.36 7.72 0.49 
Sadness 6.67 1.34 6.23 1.39 
Disgust 5.74 1.72 5.53 1.64 
Fear 6.55 1.46 6.12 1.85 
Anger 5.82 2.09 4.69 2.48 
 HC
 
PD
 
 M SD M SD 
Happiness 7.90 0.36 7.72 0.49 
Sadness 6.67 1.34 6.23 1.39 
Disgust 5.74 1.72 5.53 1.64 
Fear 6.55 1.46 6.12 1.85 
Anger 5.82 2.09 4.69 2.48 

Reaction times

We also computed the reaction times (see Table 2) in order to analyze the possible influence of response slowing in PD performance. A mixed ANOVA 2 (Group: PD patients vs. HC) × 2 (Stimuli: EFE-word and Color-word) was carried out with reaction time as the dependent measure. Neither a significant Group × Stimuli interaction (F(1,100) = 0.02, p = .89, η2 < .001), nor a significant main effect of Group (F(1,100) = 3.34, p = .07, η2 = .03) were observed. However, there was a significant main effect of Stimuli (F(1,100) = 436.31, p < .001, η2 = .81), showing that all participants were slower in EFE-word than in Color-word.

We also performed an additional analysis of the number of omitted responses due to not choosing an answer within the 10,000 ms period (see Table 2). The analysis revealed a significant main effect of Stimuli (F(1,100) = 49.68, p < .001, η2 = .33), with all participants having more omission in EFE-word than in Color-word; and a nonsignificant main effect of Group (F(1,100) = 3.05, p = .08, η2 = .03). The analysis also showed a significant interaction between Group and Stimuli (F(1,100) = 4.62, p < .05, η2 = .04). Analysis for simple effects showed a statistically significant difference when comparing PD and HC scores for EFE-word (F(1,100) = 4.13, p < .05, η2 = .04), but not for Color-word (F(1,100) = 0.14, p = .08, η2 = .003). Thus, PD patients had more omission than HC in EFE-word.

Visual Search Task

Accuracy

Similar analyses to those carried out for the Inhibition task were obtained for the Visual Search task. Firstly, a two-way mixed ANOVA was performed with Group between-subjects factor (PD patients vs. HC) and the repeated factor Stimuli (EFE crowd and Nonemotional face crowd) with accuracy scores as the dependent variable (see Table 4). The results showed a significant main effect of Stimuli (F(1,96) = 45.67, p < .001, η2 = .32), with all participants being more accurate in EFE crowd than in Nonemotional face crowd; and a significant main effect of Group (F(1,96) = 7.66, p < .05, η2 = .07), with less accuracy shown by PD patients than HC in all sets. We also observed a significant interaction between Group and Stimuli (F(1,96) = 4.56, p < .05, η2 = .05). The analysis for simple effects (Fig. 2) showed a significant difference when comparing PD and HC scores for EFE crowd (F(1,96) = 14.80, p < .001, η2 = .13) but not for Nonemotional crowd (F(1,96) = 1.21, p = .27, η2 = .01). In EFE crowd, PD patients’ mean accuracy score was below that of the HC group.

Table 4.

Mean and standard deviations for accuracy scores, reaction times and omitted responses in visual search task, both for EFE crowd and Nonemotional face crowd sets, by HC and PD patients

 Face-in-the-crowd task HC (N = 49)
 
PD (N = 49)
 
Average 
 Set M SD M SD 
Accuracy (/20) EFE crowd 16.57 1.62 14.10 2.87 15.34 
Nonemotional face crowd 13.90 2.81 12.87 2.91 13.39 
Average 15.52  14.14   
Reaction times (/10,000) EFE crowd 5,663.64 559.60 5,788.01 803.92 5,725.83 
Nonemotional face crowd 5,803.85 681.93 5,864.65 750.91 5,834.25 
Average 5,545.86  5,417.67   
Omitted responses (/20) EFE crowd 1.20 1.32 2.10 1.93 1.65 
Nonemotional face crowd 2.39 1.84 2.39 2.22 2.03 
Average 1.79  2.25   
 Face-in-the-crowd task HC (N = 49)
 
PD (N = 49)
 
Average 
 Set M SD M SD 
Accuracy (/20) EFE crowd 16.57 1.62 14.10 2.87 15.34 
Nonemotional face crowd 13.90 2.81 12.87 2.91 13.39 
Average 15.52  14.14   
Reaction times (/10,000) EFE crowd 5,663.64 559.60 5,788.01 803.92 5,725.83 
Nonemotional face crowd 5,803.85 681.93 5,864.65 750.91 5,834.25 
Average 5,545.86  5,417.67   
Omitted responses (/20) EFE crowd 1.20 1.32 2.10 1.93 1.65 
Nonemotional face crowd 2.39 1.84 2.39 2.22 2.03 
Average 1.79  2.25   

We also repeated the first analysis including GDS-R score as a covariate in order to analyze the possible influence of depression on PD performance. Results revealed that GDS-R was significant (F(1,95) = 4.39, p < .05, η2 = .05). However, we also found that the significant interaction obtained in the previous analysis was maintained (F(1,95) = 4.49, p < .05, η2 = .05), and that differences between PD and HC groups in EFE crowd were still significant (F(1,95) = 10.35, p < .01, η2 = .09).

Finally, we analyzed accuracy scores for the different EFEs (see Table 5) performing a mixed ANOVA with Group (PD patients vs. HC) as the between-subjects factor and emotion (happiness, sadness, anger, disgust, and fear) as the within-subjects factor. There was not a statistically significant interaction effect (F(4,396) = 1.04, p = .39, η2 = .01), showing that differences between PD and HC did not seem attributable to any specific emotion.

Table 5.

Mean and standard deviation for accuracy scores for each emotion in EFE crowd set, by HC and PD patients

 HC
 
PD
 
 SD SD 
Happiness 3.35 0.72 3.04 0.86 
Sadness 2.94 0.97 2.43 1.09 
Disgust 3.28 0.87 2.94 1.01 
Fear 2.45 0.58 2.20 0.74 
Anger 3.08 0.76 2.94 1.01 
 HC
 
PD
 
 SD SD 
Happiness 3.35 0.72 3.04 0.86 
Sadness 2.94 0.97 2.43 1.09 
Disgust 3.28 0.87 2.94 1.01 
Fear 2.45 0.58 2.20 0.74 
Anger 3.08 0.76 2.94 1.01 

Reaction times

To analyze possible differences between groups in speed response (see Table 4) a similar mixed ANOVA 2 (Group: PD patients vs. HC) × 2 (Stimuli: EFE and color) was carried out, with reaction time as the dependent measure. Neither a significant Group × Stimuli interaction (F(1,96) = 0.002, p = .96, η2 < .001), nor a main effect of Group (F(1,96) = 0.03, p = .86, η2 < .001) or Stimuli (F(1,96) = 0.29, p = .59, η2 = .003) was found.

As in the previous task, we also performed an additional analysis with the number of omitted responses as the dependent variable. The analysis revealed a significant main effect of Stimuli (F(1,96) = 13.04, p < .001, η2 = .12, with all participants having more omissions in Nonemotional crowd than in EFE crowd; and a nonsignificant main effect of Group (F(1,96) = 2.03, p = .16, η2 = .02). The results also showed a significant Group × Stimuli interaction (F(1,96) = 4.87, p < .05, η2 = .05), showing that PD patients had more omitted responses than HC in the EFE crowd (F(1,96) = 7.23, p < .01, η2 = .07), but not in the Nonemotional face crowd (F(1,96) = 0.00, p = 1.00, η2 = .00).

Discussion

The aim of this research is to study the relationship between selective attention and EFE recognition abilities in PD compared with healthy individuals. For this purpose, we have designed two tasks to assess different components of selective attention: Inhibition and Visual Search abilities. We have also compared both groups' responses to EFE and other nonemotional stimuli (colors and faces). This design permits us to establish whether selective attention is globally impaired in PD or, conversely, whether there is specific damage which is shown only to emotional faces. Results show that PD patients perform less well than HC, both in Inhibition and Visual Search tasks with EFE. In contrast, performance with a nonemotional stimulus is similar to HC in both tasks.

This PD performance does not seem to be related to motor slowing as we do not observe differences between the reaction times of patients and healthy individuals. Furthermore, the mean of omitted responses is greater in PD than in HC only when they see EFE, but not in the nonemotional tasks (both in Inhibition and Visual Search tasks). So, while no differences between the reaction times of both groups indicates no influence of motor slowing, in our PD results the greater number of omissions to EFE imply a particular difficulty with these kind of stimuli.

Regarding a possible influence on the results of depressive symptoms of PD individuals in spite of their higher mean scores on the depression scale administered, they do not reach the clinical criterion for depression. Moreover, we have introduced depression as a covariate in the statistical analysis to assess accuracy and the significant interaction between group and stimuli (indicating lower EFE recognition of the PD group) is maintained. Thus, differences between PD and HC, both in EFE Inhibition task and in EFE Visual Search task, are not likely to be explained by the subclinical depressive mood of PD patients. This is also in accordance with the conclusion of the meta-analysis carried out by Gray and Tickle-Degnen (2010) which point to depression as a nonrelevant factor to explain EFE recognition difficulties in PD. Some other studies have also indicated that mild to moderate depressive signs of patients do not have a relevant influence on cognitive performance (Costa, Peppe, Carlesimo, Pasqualetti, & Caltagirone, 2006; Klepac, Trkulja, & Relja, 2008).

It is also noteworthy that the patients’ poorer performance with EFEs cannot be attributed to a greater complexity of these kinds of stimuli in comparison to the others used. In this sense, the worse performance in the EFE tasks of PDs' is observed, both in the Inhibition task, when this task is more difficult than the other (color) task for all participants, and also in the Visual Search task, when the EFE task seems to be easier than the Nonemotional task. Thus, we could infer that task difficulty itself is not a suitable explanation for the worse performance of PD patients respect to HC.

Hence, our current results seem to indicate that selective attention abilities are not globally damaged in PD, and cannot alone explain the deficit observed in EFE tasks. Instead, results seem to report the presence of a specific deficit in EFE recognition among PD. These results are globally congruent to previous literature, indicating that PD patients show EFE recognition impairments for all emotions studied (Beatty et al., 1989; Breitenstein, Daum, & Ackermann, 1998; Dujardin et al., 2004; Herrera et al., 2011; Yip et al., 2003). However, other researchers have only found it in some of them (Assogna et al., 2010; Kan et al., 2002; Lachenal-Chevallet et al., 2006; Suzuki et al., 2006) and, on the whole, PD individuals tend to be more impaired at recognizing negative emotions (anger, disgust, fear, and sadness) than positive ones (Gray & Tickle-Degnen, 2010). In any case, we must emphasize that we have used a relatively reduced number of trials for each emotion, so future studies could include a greater number of trials in order to analyze the specificity versus generality of the EFE recognition impairments among PD patients.

At this point it is also necessary to explain why we do not observe deficits in nonemotional inhibition and visual search, as several studies have previously found (Djamshidian et al., 2011; Horowitz et al., 2006; Hsieh et al., 2008; Mannan et al., 2008). One plausible possibility may be that the effect of dopaminergic therapy improves global attention performance (as reflected by nonemotional stimuli). As mentioned previously, this has been suggested in some previous studies and could indicate the relevance of frontostriatal dopaminergic circuits in executive processes and, in particular, in selective attention processes (Djamshidian et al., 2011; Fera et al., 2007; Jahanshahi et al., 2000; Kehagia, Cools, Barker, & Robbins, 2009). Our PD patients were tested just after medication was administered (On state) and the fact that EFE recognition impairment persists could be related to the possible implication of other nondopaminergic circuits in emotional processing. Thus, it is known that mesolimbic cortical and subcortical brain areas play a very relevant role in emotional recognition and have been found to be damaged as the course of the disease progresses (Péron, Dondaine, Lejeune, Grandjean, & Vérin, 2012). Our results could, therefore, provide indirect evidence that damage in nondopaminergic circuits in PD is related to emotional recognition difficulties.

The lack of general cognitive impairments may also be related with the variable performance shown by our PD patients in relation to HC in all the tasks we administered. This result is congruent with the widely heterogeneous and variable pattern of cognitive performance in nondemented PD individuals (Kehagia et al., 2010). Some patients develop widespread cognitive deficits from relatively early stages, while others remain relatively intact in their abilities throughout the course of their disease (Schrag & Schott, 2006). Hence, while some authors have reported significant alterations in attention deficit in nondemented PD patients, others have found that these are only a characteristic feature in those diagnosed with dementia (Ballard et al., 2002). Moreover, among nondemented PD patients, it has been suggested that it is possible to differentiate several PD subtypes as a function of their cognitive performance (Williams-Gray, Foltynie, Brayne, Robbins, & Barker, 2007). In this same line, Foltynie, Brayne, Robbins and Barker (2004) have observed a notable variability in cognitive impairments in the disease, estimating that executive function deficit (including attention) may affect 12%, memory deficits 8%, and global deficits 15% of patients. This heterogeneity may also be related to the variable set of, mainly dopaminergic, but also cholinergic, noradrenergic, and serotonergic deficits that characterize PD (Kehagia et al., 2010).

Nevertheless, it should be taken into account that we have employed two newly developed experimental tasks which may be not wholly comparable with the traditional Stroop and Face-in-the-Crowd paradigms, and this may limit the explanation we suggest. In any case, even considering possible differences in stimuli, our present results show a specific deficit in EFE recognition which is not dependent on attention abilities. This impairment may have clinical relevance because the recognition of other people's faces plays a central role in social communication, and is one of the most complex perceptive abilities in humans (Haxby & Gobbini, 2011). Different information can be extracted from an individual face, including identity, age, gender, the direction of gaze and, of course, the emotional state of others through EFE. They are especially relevant to guide social activities and, hence, recognition problems may be a potential source for communicative difficulties and social isolation in PD individuals. For example, Clark and colleagues (2008) have shown that their emotion recognition disabilities are correlated with different subjective complaints, such as frustration in social relations, feelings of social disconnection, and a desire to connect with others. These interpersonal communication difficulties may have a great impact on patients' quality of life, and some studies have indicated that they can be even more detrimental than their physical symptoms (Schreurs, De Ridder, & Bensing, 2000). So, from a clinical point of view, our results emphasize the need for a specific assessment and rehabilitation of emotional recognition abilities, as a way to prevent possible problems in these communicative skills. As our results indicate, these problems may be subtle and independent of cognitive performance in PD individuals. Nonetheless, we are aware that our results may be contrasted and confirmed with complementary works, including, for example, an assessment of emotional recognition abilities in more natural settings and, as suggested previously, a more in-depth examination of all the basic emotions.

Funding

This work was supported by the Ministry of Science and Innovation (PSI2009-09067).

Acknowledgements

We thank the Asociación Parkinson Alcorcón y Otros Municipios (APARKAM, Alcorcón, Madrid, Spain), the Asociación Parkinson Madrid (Madrid, Spain), the Asociación Parkinson San Sebastián de los Reyes (ALCOSSE, San Sebastián de los Reyes, Madrid, Spain), and the Centro de Mayores el Soto (Móstoles, Madrid, Spain).

References

Adolphs
R.
Recognizing emotion from facial expressions: Psychological and neurological mechanisms
Behavioral and Cognitive Neuroscience Review
 
2002
1
21
62
Alonso-Recio
L.
Martín
P.
Carvajal
F.
Ruiz
M.
Serrano
J. M.
A holistic analysis of relationship between executive function and memory in Parkinson's disease
Journal of Clinical and Experimental Neuropsychology
 
2013
35
2
147
159
Alonso-Recio
L.
Martín-Plasencia
P.
Loeches-Alonso
Á.
Serrano-Rodríguez
J. M.
Working memory and facial expression recognition in patients with Parkinson's disease
Journal of the International Neuropsychological Society
 
2014
20
1
10
Alonso-Recio
L.
Martín
P.
Rubio
S.
Serrano
J. M.
Discrimination and categorization of emotional facial expressions and faces in Parkinson's disease
Journal of Neuropsychology
 
2013
Advance online publication doi:10.1111/jnp.12029
Alonso-Recio
L.
Serrano-Rodriguez
J. M.
Carvajal-Molina
F.
Loeches-Alonso
A.
Martin-Plasencia
P.
Recognition of facial expression of emotions in parkinson's disease: A theoretical review [Reconocimiento de expresiones faciales de emociones en la enfermedad de Parkinson: Una revision teórica]
Revista de Neurología
 
2012
54
8
479
489
Ariatti
A.
Benuzzi
F.
Nichelli
P.
Recognition of emotions from visual and prosodic cues in parkinson's disease
Neurological Sciences
 
2008
29
4
219
227
Assogna
F.
Pontieri
F. E.
Caltagirone
C.
Spalleta
G.
The recognition of facial emotion expressions in Parkinson's disease
European Neuropsychopharmacology
 
2008
18
11
835
848
Assogna
F.
Pontieri
F. E.
Cravello
L.
Peppe
A.
Pierantozzi
M.
Stefani
A.
et al.  
Intensity-dependent facial emotion recognition and cognitive functions in parkinson's disease
Journal of the International Neuropsychology Society
 
2010
16
5
867
876
Ballard
C. G.
Aarsland
D.
McKeith
I.
ÓBrien
J.
Gray
A.
Cormack
F.
et al.  
Fluctuations in attention: PD dementia vs DLB with parkinsonism
Neurology
 
2002
59
11
1714
1720
Barone
P.
Aarsland
D.
Burn
D.
Emre
M.
Kulisevsky
J.
Weintraub
D.
Cognitive impairment in nondemented parkinson's disease
Movement Disorders
 
2011
26
14
2483
2495
Beatty
W. W.
Goodkin
D. E.
Weir
W. S.
Staton
R. S.
Monson
N.
Beatty
P. A.
Affective judgements by patients with Parkinson's disease and chronic progressive multiple sclerosis
Bulletin of the Psychonimic Society
 
1989
27
4
361
364
Breitenstein
C.
Daum
I.
Ackermann
H.
Emotional processing following cortical and subcortical brain damage: Contribution of the fronto-striatal circuitry
Behavioral Neurology
 
1998
11
1
29
42
Chaudhuri
K. R.
Healy
D. G.
Schapira
A. H.
Non- motor symptoms of Parkinson's disease: Diagnosis and management
Lancet Neurology
 
2006
5
235
245
Clark
U. S.
Neargarder
S.
Cronin-Golomb
A.
Specific impairments in the recognition of emotional facial expressions in parkinson's disease
Neuropsychologia
 
2008
46
9
2300
2309
Cohen
H.
Gagne
M. H.
Hess
U.
Pourcher
E.
Emotion and object processing in parkinson's disease
Brain and Cognition
 
2010
72
3
457
463
Costa
A.
Peppe
A.
Carlesimo
G. A.
Pasqualetti
P.
Caltagirone
C.
Major and minor depression in Parkinson's disease: A neuropsychological investigation
European Journal of Neurology
 
2006
13
972
980
Dara
C.
Monetta
L.
Pell
M. D.
Vocal emotion processing in Parkinson's disease: Reduced sensitivity to negative emotions
Brain Research
 
2008
1188
100
111
Djamshidian
A.
O'Sullivan
S. S.
Lees
A.
Averbeck
B. B.
Stroop test performance in impulsive and non impulsive patients with Parkinson's disease
Parkinsonism & Related Disorders
 
2011
17
3
212
214
Dujardin
K.
Blairy
S.
Defebvre
L.
Duhem
S.
Noel
Y.
Hess
U.
et al.  
Deficits in decoding emotional facial expressions in parkinson's disease
Neuropsychologia
 
2004
42
2
239
250
Dujardin
K.
Tard
C.
Duhamel
A.
Delval
A.
Moreau
C.
Devos
D.
et al.  
The pattern of attentional deficits in Parkinson's disease
Parkinsonism & Related Disorders
 
2013
19
3
300
305
Ebner
N. C.
Riediger
M.
Linderberger
U.
FACES—A database of facial expressions in young, middle, and older women and men: Development and validation
Behavior Research Methods
 
2010
42
1
351
362
Fera
F.
Nicoletti
G.
Cerasa
A.
Romeo
N.
Gallo
O.
Gioia
M. C.
et al.  
Dopaminergic modulation of cognitive interference after pharmacological washout in Parkinson's disease
Brain Research Bulletin
 
2007
74
75
83
Folstein
M. F.
Folstein
S. E.
McHugh
P. R.
‘Mini-mental state’. A practical method forgrading the cognitive state of patients for the clinician
Journal of Psychiatric Research
 
1975
12
189
198
Foltynie
T.
Brayne
C. E.
Robbins
T. W.
Barker
R. A.
The cognitive ability in an incident cohort of Parkinson's patients in the UK. The CamPaIGN study
Brain
 
2004
127
550
560
García-Rodríguez
B.
Casares Guillén
C.
Jurado Barba
R.
Rubio-Valladolid
G.
Molina Arjona
J. A.
Ellgring
H.
Visuo-spatial interference affects the identification of emotional facial expressions in unmedicated Parkinson's patients
Journal of the Neurological Sciences
 
2012
313
13
16
Gray
H. M.
Tickle-Degnen
L.
A meta-analysis of performance on emotion recognition tasks in parkinson's disease
Neuropsychology
 
2010
24
2
176
191
Hammond
C.
Bergman
H.
Brown
P.
Pathological synchronization in Parkinson's disease: Networks, models and treatments
Trends in Neurosciences
 
2007
30
7
357
364
Hansen
C. H.
Hansen
R. D.
Finding the face in the crowd: An anger superiority effect
Journal of Personality and Social Psychology
 
1988
54
6
917
924
Haxby
J.V.
Gobbini
M.I.
Rhodes
G.
Calder
A.
Johnson
M.
Haxby
J.V.
Distributed neural systems for face perception
Oxford Handbook of Face Perception
 
2011
93
110
Oxford: Oxford University Press
Herrera
E.
Cuetos
F.
Rodriguez-Ferreiro
J.
Emotion recognition impairment in parkinson's disease patients without dementia
Journal of the Neurological Sciences
 
2011
310
1–2
237
240
Hoehn
M. M.
Yahr
M. D.
Parkinsonism: Onset, progression and mortality
Neurology
 
1967
17
427
442
Horowitz
T. S.
Choi
W. Y.
Horvitz
J. C.
Côté
L. J.
Mangels
J. A.
Visual search deficits in Parkinson's disease are attenuated by botton-up salience and top-down information
Neuropsychologia
 
2006
44
1962
1977
Hsieh
Y. H.
Chen
K. J.
Wang
C. C.
Lai
C. L.
Cognitive and motor components of response speed in the Stroop test in Parkinson's disease patients
The Kaohsiung Journal of Medical Sciences
 
2008
24
4
197
203
Hughes
A. J.
Ben-Shlomo
Y.
Daniel
S. E.
Lees
A. J.
What features improve the accuracy of clinical diagnosis in Parkinson's disease: A clinicopathologic study. 1992
Neurology
 
2001
57
10 Suppl. 3
S34
S38
Izal
M.
Montorio
I.
Nuevo
R.
Perez-Rojo
G.
Cabrera
I.
Optimising the diagnostic performance of the Geriatric Depression Scale
Psychiatry Research
 
2010
178
142
146
Jahanshahi
M.
Ardouin
C. M. A.
Brown
R. G.
Rothwell
J. C.
Obeso
J.
Albanese
A.
et al.  
The impact of brain stimulation on executive function in Parkinson's disease
Brain
 
2000
123
1142
1154
Jankovic
J.
Parkinson's disease: Clinical features and diagnosis
Journal of Neurology, Neurosurgery and Psychiatry
 
2008
79
368
376
Kan
Y.
Kawamura
M.
Hasegawa
Y.
Mochizuki
S.
Nakamura
K.
Recognition of emotion from facial, prosodic and written verbal stimuli in parkinson's disease
Cortex
 
2002
38
4
623
630
Kehagia
A. A.
Barker
R. A.
Robbins
T. W.
Neuropsychological and clinical heterogeneity of cognitive impairment and dementia in patients with parkinson's disease
Lancet Neurol
 
2010
9
12
1200
1213
Kehagia
A. A.
Cools
R.
Barker
R. A.
Robbins
T. W.
Switching between abstract rules reflects disease severity but not dopaminergic status in Parkinson's disease
Neuropsychologia
 
2009
47
1117
1127
Klepac
N.
Trkulja
V.
Relja
M.
Nondemented Parkinson disease patients: Is cognitive performance associated with depressive difficulties?
Cognitive and Behavioral Neurology
 
2008
21
87
91
Lachenal-Chevallet
K.
Bediou
B.
Bouvard
M.
Thobois
S.
Broussolle
E.
Vighetto
A.
et al.  
Emotional facial expression recognition impairment in parkinson disease
Psychol Neuropsychiatr Vieil
 
2006
4
1
61
67
Lezak
M. D.
Howieson
D. B.
Loring
D. W.
Neuropsychological assessment (4th ed.)
 
2004
New York
Oxford University Press
Mannan
S. K.
Hodgson
T. L.
Husain
M.
Kennard
C.
Eye movements in visual search indicate impaired saliency processing in Parkinson's disease
Progress in Brain Research
 
2008
171
559
562
Martinez-Espinosa
S.
Delgado-Villapalos
C.
Bermejo
F.
Estimation of premorbid intelligence in Spanish people with the Word AccentuationTest and its application to the diagnosis of dementia
Brain and Cognition
 
1997
33
343
356
Martinez-Martin
P.
Forjaz
M. J.
Cubo
E.
Frades
B.
de Pedro Cuesta
J.
Global versus factor-related impression of severity in Parkinson's disease: A new clinimetric index (CISI-PD)
Movement Disorders
 
2006
21
208
214
Martinez-Martin
P.
Frades Payo
B.
Quality of life in Parkinson's disease: Validation study of the PDQ-39 Spanish version. The Grupo Centro for Study of Movement Disorders
Journal of Neurology
 
1998
245
Suppl. 1
S34
S38
Merello
M.
Trastornos no motores en la enfermedad de Parkinson
Revista de Neurología
 
2008
47
5
261
270
Muslimovic
D.
Post
B.
Speelman
J. D.
De Haan
R. J.
Schmand
B.
Cognitive decline in Parkinson's disease: A prospective longitudinal study
Journal of the International Neuropsychological Society
 
2009
15
426
437
Pell
M. D.
Leonard
C. L.
Facial expression decoding in early parkinson's disease
Cognitive Brain Research
 
2005
23
2–3
327
340
Péron
J.
Dondaine
T.
Le Jeune
F.
Grandjean
D.
Verin
M.
Emotional processing in parkinson's disease: A systematic review
Movement Disorders
 
2012
27
2
186
199
Schneider
W.
Eschman
A.
Zuccolotto
A.
E-prime user's guide
 
2002
Pittsburg, CA
Psychology Software Tools Inc
Schrag
A.
Schott
J. M.
Epidemiological, clinical and genetic characteristics of early-onset parkinsonism
Lancet Neurology
 
2006
5
355
363
Schreurs
K. M. G.
De Ridder
D. T. D.
Bensing
J. M.
A one year study of coping, social support and quality of life on Parkinson's disease
Psychology & Health
 
2000
15
109
121
Schröder
C.
Mobes
J.
Schutze
M.
Szymanowski
F.
Nager
W.
Bangert
M.
et al.  
Perception of emotional speech in parkinson's disease
Movement Disorders
 
2006
21
10
1774
1778
Schwab
R. S.
England
A. C.
Gillingham
F. J.
Donaldson
I. M. L.
Projection technique for evaluating surgery in Parkinson_s disease
Third symposium on Parkinson's disease
 
1969
Edinburgh, UK
E. & S. Livingstone
152
157
Simons
G.
Ellgring
H.
Pasqualini
M. C.
Disturbance of spontaneous and posed facial expressions in Parkinson's disease
Cognition and Emotion
 
2003
17
759
778
Sohlberg
M. M.
Mateer
C. A.
Effectiveness of an attention-training program
Journal of Clinical and Experimental Neuropsychology
 
1987
9
117
130
Stroop
J. R.
Studies of interference in serial verbal reactions
Journal of Experimental Psychology
 
1935
18
6
643
662
Suzuki
A.
Hoshino
T.
Shigemasu
K.
Kawamura
M.
Disgust-specific impairment of facial expression recognition in parkinson's disease
Brain
 
2006
129
Pt 3
707
717
Vuilleumier
P.
Facial expression and selective attention
Current Opinion in Psychiatry
 
2002
15
291
300
Williams-Gray
C. H.
Foltynie
T.
Brayne
C. E. G.
Robbins
T. W.
Barker
R. A.
Evolution of cognitive dysfunction in an incident Parkinson's disease cohort
Brain
 
2007
130
7
1787
1798
Yip
J. T.
Lee
T. M.
Ho
S. L.
Tsang
K. L.
Li
L. S.
Emotion recognition in patients with idiopathic parkinson's disease
Movement Disorders
 
2003
18
10
1115
1122