Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2023 Aug 1.
Published in final edited form as: Neuropsychol Rehabil. 2021 Feb 21:1–19. doi: 10.1080/09602011.2021.1888756

Emotion recognition impairments and social well-being following right-hemisphere stroke

Katherine O’Connell 1,*, Abigail A Marsh 2, Dorothy Farrar Edwards 3, Alexander W Dromerick 4,5, Anna Seydell-Greenwald 5
PMCID: PMC8379297  NIHMSID: NIHMS1675568  PMID: 33615994

Abstract

Accurately recognizing and responding to the emotions of others is essential for proper social communication and helps bind strong relationships that are particularly important for stroke survivors. Emotion recognition typically engages cortical areas that are predominantly right-lateralized including superior temporal and inferior frontal gyri—regions frequently impacted by right-hemisphere stroke. Since prior work already links right-hemisphere stroke to deficits in emotion recognition, this research aims to extend these findings to determine whether impaired emotion recognition after right-hemisphere stroke is associated with worse social well-being outcomes. Eighteen right-hemisphere stroke patients (≥6 months post-stroke) and 21 neurologically healthy controls completed a multimodal emotion recognition test (Geneva Emotion Recognition Test – Short) and reported engagement in social/non-social activities and levels of social support. Right-hemisphere stroke was associated with worse emotion recognition accuracy, though not all patients exhibited impairment. In line with hypotheses, emotion recognition impairments were associated with greater loss of social activities after stroke, an effect that could not be attributed to stroke severity or loss of non-social activities. Impairments were also linked to reduced patient-reported social support. Results implicate emotion recognition difficulties as a potential antecedent of social withdrawal after stroke and warrant future research to test emotion recognition training post-stroke.

Keywords: stroke, right-hemisphere, emotion recognition, social activity, social support

Introduction

The ability to discriminate whether someone is loudly exclaiming in joy or in anger relies on our capacity for emotion recognition. In real-world interactions, we integrate rich visual information from facial expressions and body language with auditory cues like emotional prosody (i.e., variations in vocal pitch, loudness and/or tempo that correspond to a speaker’s emotional state) to contextualize and interpret others’ emotions. Damage to brain regions involved in emotion perception, for example after a stroke, may hinder the ability to recognize these cues of social and emotional communication. As a result, individuals with impaired emotion recognition may find everyday conversations cumbersome or confusing, while individuals in their social network may feel frustrated or discouraged by being misunderstood. Such outcomes have important clinical implications. For stroke survivors in particular, strong and supportive social networks are associated with better rehabilitation compliance (Evans et al., 1987) and improved long-term clinical outcomes (Beckley, 2006; Glass et al., 1993; Lehnerer et al., 2019) including reduced morbidity and mortality (Colantonio et al., 1993; Dupre & Lopes, 2016; Northcott et al., 2016; Vogt et al., 1992). Better understanding the effects of stroke on emotion recognition and social well-being—and how deficits in these domains can be rehabilitated—is essential given that the number of stroke survivors is projected to increase over the next decade. By 2030 an estimated 3.9% of U.S. adults will have suffered a stroke (Ovbiagele et al., 2013), a 22% increase relative to the 3.2% reported in 2019 (Centers for Disease Control and Prevention, 2019).

Whereas language and semantic processing are predominantly left-lateralized for most people, core aspects of emotion processing are thought to be right-lateralized (Adolphs et al., 1996, 2000; Borod et al., 1998; Gorelick & Ross, 1987; Heilman et al., 1975; Ross & Mesulam, 1979; Ross & Monnot, 2008; Seydell-Greenwald et al., 2020; Tucker et al., 1977; Weintraub et al., 1981). Work in neurologically healthy adults indicates the right hemisphere receives facial expression information earlier (Streit et al., 1999) and shows larger, more reliable responses to emotional faces than the left hemisphere (Adolphs et al., 1996; De Winter et al., 2015; Dzhelyova et al., 2017; Kesler-West et al., 2001; Leleu et al., 2019; Sliwinska & Pitcher, 2018). Transcranial stimulation disrupting right frontotemporal areas acutely impairs facial emotion processing in healthy adults (Chick et al., 2020; Pitcher, 2014), and one study showed significantly greater disruption for stimulation applied to right relative to left posterior superior temporal sulcus (Sliwinska & Pitcher, 2018). For vocal emotion processing (i.e., emotional prosody), a meta-analysis of 27 studies found associated activation across right and left frontotemporal areas, but a formal lateralization analysis suggested right-dominant specialization for emotion in temporal gyri (Witteman et al., 2012). These findings all suggest a right-hemisphere lateralization for emotion processing and provide evidence against the chief alternative theory, the valence hypothesis, which instead posits that right- versus left-hemispheric involvement depends on emotional valence (Canli et al., 1998; Davidson, 1995).

A right-hemisphere preferential bias for emotion perception aids in explaining the socioemotional impacts often observed after right-hemisphere stroke. A 2013 meta-analysis reported that of 35 published papers investigating facial emotion recognition after stroke, 32 (91%) found impaired recognition accuracy in patients with right-hemisphere stroke compared to only 1 (3%) paper that found impairment after left-hemisphere stroke (Yuvaraj et al., 2013). A 2011 meta-analysis of emotional prosody found that a stroke injury to either hemisphere was associated with compromised recognition accuracy, but the effect of right-hemisphere damage was significantly greater (g=0.37, n=620; Witteman et al., 2011). Recent work in acute right-hemisphere stroke patients revealed that greater damage to right posterior superior temporal gyrus significantly predicted emotional prosody deficits (Sheppard et al., 2020), and in a small sub-sample, these prosody impairments persisted for at least 6 months post-stroke in the majority of acutely impaired patients. To our knowledge, only two prior studies have investigated the potential real-world implications of emotion recognition deficits after stroke. Cooper and colleagues reported that across patients with bilateral, right-hemisphere and left-hemisphere stroke, emotion recognition accuracy was associated with restrictions in social participation in a U.K. sample (Cooper et al., 2014). Stiekema and colleagues, however, reported no similar association in a Dutch sample (Stiekema et al., 2019). These inconsistencies incite motivation for the present research, which focuses on right-hemisphere stroke.

The current work aims to assess how emotion recognition deficits after stroke are associated with real-world outcomes related to social well-being. Two facets of social well-being are considered: changes to social activity, and levels of social support. Here, social activities refer to behaviors that involve interaction with others (for example, talking on the telephone or visiting friends) while social support reflects the quantity and quality of social connections. We focus on patients who are most likely to experience long-term impacts of emotion recognition deficits—right-hemisphere stroke survivors in the chronic phase of stroke. Emotion recognition accuracy was measured in these patients and demographically-matched controls using a task that simultaneously incorporated dynamic vocal, facial and body emotional cues analogous to real-life emotional communication (Schlegel et al., 2014) to enhance ecological validity. If emotion recognition difficulties after right-hemisphere stroke compromise social communication abilities, then patients with worse emotion recognition accuracy are expected to have: 1) reduced social activity (due to withdrawal or exclusion) and, 2) lower social support (due to a loss of social connections). These hypotheses were tested using adjustment for covariates including stroke severity and change to non-social activity.

Methods

Participants

We enrolled 22 right-hemisphere stroke patients in the chronic (≥ 6 months) phase of stroke recovery and 31 demographically-matched healthy control subjects, all over the age of 45. Participants were recruited from the Washington, DC Metro Area between 2017 and 2019. The final sample excluded patients who had bilateral stroke involvement (n=2), incomplete data (n=1) or significant comprehension difficulties (n=1). Control subjects were excluded for abnormal MRI (n=2), medical conditions with neurological involvement (n=2), Montreal Cognitive Assessment (MoCA) score < 20 (n=4) (Waldron-Perrine & Axelrod, 2012) or incomplete data (n=2).

The final sample included 18 right-hemisphere patients (11M, 7F) and 21 control subjects (13M, 8F). Demographic characteristics of the final sample are reported in Table 1. All study procedures were carried out in accordance with a protocol approved by the Institutional Review Board at Georgetown University in Washington, DC, and subjects provided written informed consent prior to testing.

Table 1.

Characteristics of the final sample

RH Stroke Control p-value[a]
N 18 21
Age, M (SD) 62.3 (9.4) 59.0 (7.2) .232
Gender .959
 Male 11 13
 Female 7 8
Race .967
 White/ Caucasian 7 9
 Black/ African-American 10 11
 Asian 1 1
Marital Status[b] .010
 Married 11 4
 Unmarried 7 17
Hollingshead Index[c], M (SD) 49.7 (16.8) 50.6 (13.4) .968
Months since stroke, M (SD) 18.8 (11.6) - -
NIHSS, M (SD) 3.9 (2.8) 0.1 (0.3) <.001
 Motor Arm 0/1/≥2 9/6/4 21/0/0 <.001
 Motor Leg 0/1/≥2 10/7/1 21/0/0 .001
Open in a new tab

Note.

[a]

P-values for continuous variables were calculated using a two-tailed t-test. P-values were obtained using a Fisher’s exact test for race and marital status and a chi-squared test for gender.

[b]

For reference, 65.7% of adults between the ages of 45-65 report being currently married in the Washington, DC Metro area (U.S. Census Bureau, 2018);

[c]

The Hollingshead Index is a measure of socioeconomic status; higher scores indicate higher socioeconomic status (range 8 to 66).

Imaging data (CT or MRI) were available for all participants and reviewed by a board-certified neurologist. All patients had ischemic strokes within the right middle cerebral artery (MCA) territory; a large intraparenchymal hemorrhagic conversion was noted in one. Additional small lesions outside the right MCA-territory (e.g., in the pons) were noted in four of the patients. All stroke patients and control subjects completed the National Institutes of Health Stroke Scale (NIHSS) assessment (National Institute of Neurological Disorders and Stroke, 2011).

Geneva Emotion Recognition Test – Short (GERT-S)

The GERT-S (Schlegel & Scherer, 2016) presents 42 brief (1-3 s) videos with sound in which actors are speaking in a pseudo language and expressing 1 of 14 emotions through tone of voice, body language and facial expression. After presentation of each video, subjects selected the emotion they thought the individual was expressing using an emotion wheel that organized emotions by valence (see Supplemental Figure 1). Participants completed the task on a computerized tablet and responded with a stylus. Three practice trials were included prior to beginning the task. No limit was imposed for response time.

The task was coded in PsychoPy 1.81.1 (Peirce, 2007), a software package designed for stimulus presentation. Overall accuracy was assessed as the number of total correct trials divided by the total number of trials completed. One patient completed 37 of the 42 trials due to a computer problem, but was included in the final sample. The GERT-S shows high internal consistency (alpha=.80-.83) and positively correlates with other measures of emotion recognition and emotion understanding (Schlegel et al., 2017; Schlegel & Scherer, 2016).

Activity Card Sort (ACS)

The ACS (Baum & Edwards, 2001, 2008) probes engagement in 89 different activities (e.g., “eating at a restaurant”, “paying bills”) through manual sorting of individual activity cards. Moderate-to-high internal consistency is observed for the four subscales, which categorize activity types (social: alpha=.77-.80; low-physical-demand leisure: alpha=.66-.71; high-physical-demand leisure: alpha=.61-.85; instrumental: alpha=.71-.83; Everard et al., 2000; Katz et al., 2003). Examples of social items include, “talking on the phone,” “visiting friends,” and “volunteer work.”

The ACS has been used to estimate changes in activity participation over time and the impact of various medical morbidities, including stroke (Chan et al., 2006; Edwards et al., 2006; Hartman-Maeir et al., 2007; Katz et al., 2003; Lyons et al., 2010). Stroke patients were asked to indicate how their current engagement in each activity differed relative to just before their stroke. Control participants were asked to indicate how their current engagement in each activity differed from a couple of years prior so as to match the time since stroke in the stroke patients. Response options included: never did, continue to do now, doing less, given up, and started this. Retention of activity level is the percentage of activities in which a person is currently involved in relative to prior engagement. This score is obtained by dividing the sum of current activities by the sum of previous activities and multiplying by 100 (“doing less” was assigned as previous=1 current=0.5). Higher scores indicate greater retention of pre-stroke/ prior activity.

Older Americans Resources and Services (OARS) Social Resources Scale

The Social Resources Scale of the OARS Multidimensional Functional Assessment contains 11 items and was developed for use in community-based, older Americans (Fillenbaum, 2013). Each item provides labeled response options. Example items include, “About how many times in a normal week do you talk to someone—friends, relatives, or other persons—on the telephone?” with responses scored as “not at all”=0, “once”=1, “2-6 times”=2, “once a day or more”=3 and, “Do you find yourself feeling lonely?” with responses scored as “almost never”=2, “sometimes”=1, “quite often”=0. The scale was administered in an interview format and responses were summed to obtain a total summary score (higher scores indicate greater support). This measure reflects multiple aspects of social support including the amount of contact with others, the availability of support, and the perceived quality of support. Note this measure assessed current levels of social support and did not inquire about changes since the stroke for patients. No other OARS scales were administered.

Data analysis

Statistical analyses were completed in Stata 15 (StataCorp. 2017. College Station, TX). Fisher’s exact tests assessed group differences on race and marital status and a chi-squared test assessed differences in gender. Two-sample t-tests were used to compare groups on emotion recognition accuracy and the Cohen’s d effect size along with corresponding 95% CI are reported. For approximately normally distributed variables, Pearson correlation tests assessed the relationship between emotion recognition accuracy and social outcomes (ACS activity retention). For non-normally distributed variables (OARS social support), Spearman rank correlation tests were used instead. All p-values represent significance of a two-tailed test. Hierarchical multiple linear regression models were used to further assess correlational relationships and adjust for covariates while testing for group effect interactions as described below.

Results

Emotion recognition accuracy

Relative to the control group, right-hemisphere stroke patients had worse overall emotion recognition accuracy (RH: M=38.3%, SD=17.3%; Control: M=49.1%, SD=14.0%; t(37)=2.16, p=.037; d=.69, 95% CI=[.04 to 1.34]; Figure 1). Similar trends were observed when assessing overall accuracy for positive emotion target trials (RH: M=47.5%, SD=20.4%; Control: M=59.3%, SD=19.0%; t(37)=1.87, p=.070, d=.60, 95% CI=[-.05 to 1.24]) and overall accuracy for negative emotion target trials (RH: M=32.9%, SD=19.6%; Control: M=42.9%, SD=16.1%; t(37)=1.74, p=.090, d=.56, 95% CI=[-.09 to 1.20]), suggesting that impairments were not limited to a particular valence.

Figure 1: Overall emotion recognition accuracy.

Figure 1:

Open in a new tab

Right-hemisphere stroke patients showed worse emotion recognition performance relative to control subjects (p=.037). Chance accuracy is 7.1%; Error bars represent 95% CI of the mean.

To determine whether minor mistakes involving mix-ups between similar emotions (e.g., mislabeling fear as anxiety or vice-versa) could explain group differences, we also assessed overall accuracy when permitting emotions with similar meanings to be interchangeable as follows: anger = irritation, sadness = despair, fear = anxiety, joy = amusement, and pleasure = relief. On average, these minor mistakes accounted for 22.9% of errors in the stroke patients and 19.4% of errors in controls; however, a group comparison revealed the between-group difference in emotion recognition accuracy persisted after coding these minor errors as correct (RH: M=49.2%, SD=18.3%; Control: M=60.2%, SD=13.6%; t(37)=2.16, p=.037, d=.69, 95% CI=[.04 to 1.34]). This result indicates that stroke patients’ worse performance could not be attributed to trivial labeling errors between similar emotions.

To visualize the nature of errors in each group on the emotion recognition task, confusion matrices display how participants responded to each target emotion (Figure 2). These matrices indicate that overall, the groups made similar mistakes; for example, both stroke patients and control participants commonly mistook fear as anger and irritation as interest. However, the right-hemisphere stroke patients had a higher proportion of responses off the diagonal, indicating lower overall accuracy (the proportion of responses off the diagonal is equivalent to 100% - overall percent accuracy).

Figure 2: Emotion recognition performance.

Figure 2:

Open in a new tab

Confusion matrices depict mean responses for each group on the GERT-S. Darker red on the diagonal indicates higher accuracy, while darker red off the diagonal indicates errors. Plots reveal that right-hemisphere stroke patients and control subjects made similar mistakes (i.e., incorrectly labeling fear stimuli as anger), but overall performance was worse for patients.

Emotion recognition accuracy in relation to social well-being

To test whether emotion recognition deficits after right-hemisphere stroke were associated with worse social well-being outcomes, we next evaluated correlations between emotion recognition accuracy and two measures of social well-being: retention of social activities and levels of social support.

A positive correlation between emotion recognition accuracy and retention of social activities was observed in the right-hemisphere stroke group (r=.506, p=.032). This relationship persisted when controlling for stroke severity using NIHSS total score (partial r=.485, p=.048). No significant relationship between emotion recognition accuracy and social activity retention (which, again, measures changes in social activity) was observed in the control group (r=−.071, p=.760).

To more fully account for the role of potential confounding variables, we investigated this association using hierarchical multiple linear regression including all participants and set emotion recognition accuracy as the dependent variable (Table 2). The model included covariates for age, sex and retention of low-demand leisure activities (which requires comparable levels of physical demand to the social activity variable and therefore adjusts for potential changes to non-social activity). A group x social activity retention interaction term was tested to examine whether for stroke patients relative to controls, poor social activity retention was associated with worse emotion recognition performance. Group was coded as 0=control and 1=right-hemisphere stroke, sex was coded as 0=male and 1=female and age was mean centered. The interaction term improved the fit of the model (ΔR2=.09, p=.025) and revealed a significantly greater positive association between retention of social activities and emotion recognition accuracy in right-hemisphere stroke patients than controls (b=56.64, 95% CI=[7.49, 105.80], p=.025; Figure 3a). In other words, after controlling for age, sex, and change to non-social leisure activities, emotion recognition impairments were associated with reduced social activity retention in the stroke patients but not controls. Consistent with prior research (Ruffman et al., 2008; Schlegel et al., 2014; Schlegel & Scherer, 2016), the model also revealed that older age predicted reduced emotion recognition accuracy (b=−.87, 95%CI=[−1.43, −0.30], p=.004) such that a 10 year increase in age was associated with a decrease of 8.7 percentage points.

Table 2.

Multiple linear regression predicting emotion recognition accuracy

Step 1 Step 2
b 95% CI p b 95% CI p
Constant 24.87 −4.00, 53.75 54.32 17.06, 91.58
Age −0.79 −1.39, -0.20 .011* −0.87 −1.43, -0.30 .004**
Sex 2.66 −6.92, 12.24 .577 2.22 −6.78, 11.23 .618
Low-leisure retention 29.27 −3.72, 62.25 .080 28.19 −2.80, 59.17 .073
Social retention −3.85 −35.99, 28.29 .809 −37.47 −79.45, 4.50 .078
Group −6.13 −15.60, 3.34 .197 −52.25 −93.24, -11.26 .014*
Social retention * Group 56.64 7.49, 105.80 .025*
R2 .374** .466**
ΔR2 .092*
Open in a new tab

Note. N = 39. Accuracy is percent correct (between 0 and 100). b represents unstandardized coefficients.

*

p<.05,

**

p<.01.

Figure 3: Emotion recognition performance is positively associated with social well-being after right-hemisphere stroke.

Figure 3:

Open in a new tab

Significant positive correlations between emotion recognition accuracy and (a) retention of social activity (RH: r=.506, p=.032), and (b) self-reported social support (RH: rho=.480, p=.044) in right-hemisphere stroke patients. Analogous relationships were not observed in the control sample (ps>.519).

Additional information about group-level responses to the ACS can be found in Supplemental Materials (Supplemental Figures 2–5).

We then tested whether emotion recognition impairments in the stroke group were associated with reduced social support as measured by the OARS Social Resources Scale. Emotion recognition accuracy significantly positively correlated with social support in right-hemisphere stroke patients (rho=.480, p=.044; Figure 3b). This correlation appears to be driven by OARS items assessing the quantity rather than the perceived quality of social contacts (Supplemental Tables 1–2). No such correlation was evident in controls (rho=.149, p=.519). We again applied hierarchical multiple linear regression including all participants, set emotion recognition accuracy as the dependent variable, and included covariates for age and sex. The model revealed only a trending association between social support and emotion recognition accuracy (b=1.38, 95%CI=[−0.24, 3.00], p=.093; Supplemental Table 3) and including a group x social support interaction term did not significantly improve model fit (ΔR2=.02, p=.362) indicating no significant difference in the relationship between social support and emotion recognition between groups.

Discussion

This research provides insight on the socioemotional consequences and correlates of right-hemisphere stroke. Findings supported our hypothesis that emotion recognition impairments after right-hemisphere stroke are associated with worse social well-being outcomes, even after adjustment for covariates such as stroke severity and changes to non-social activities. Patients showed worse performance than demographically-matched controls on a dynamic, multimodal emotion recognition task. Levels of impairment in the stroke group, but not controls, were associated with reduced social activity engagement suggesting that emotion recognition deficits may play a role in patients becoming isolated despite widespread consensus of the importance of maintaining social connections in stroke recovery (Boden-Albala et al., 2005; Colantonio et al., 1993; Glass et al., 1993; Glass & Maddox, 1992; Nijsse et al., 2019; Northcott et al., 2016; Shimoda & Robinson, 1998; Wellman & Wortley, 1990). Specifically, emotion recognition impairments were associated with greater loss of social activities, and this relationship persisted after adjusting for stroke severity and loss of non-social activities, indicating the effect is not driven by other stroke sequelae (e.g., hemiparesis). A relationship between emotion recognition impairments and reduced patient-reported social support was also observed, though regression modeling indicated this relationship did not significantly differ between patients and controls. Due to the clinical relevance of social well-being after stroke, for example in supportive networks helping patients adhere to rehabilitation regimens (Evans et al., 1987) and improving long-term recovery outlook (Beckley, 2006; Glass et al., 1993; Lehnerer et al., 2019; Vogt et al., 1992), these findings warrant future work to test whether alleviating emotion recognition deficits beneficially impacts social participation and rehabilitation after right-hemisphere stroke.

These findings corroborate a small but growing literature reporting associations between emotion recognition difficulties and social well-being after neurological insults including stroke (Cooper et al., 2014; but see Stiekema et al., 2019) and traumatic brain injury (Knox & Douglas, 2009; McDonald et al., 2004; Spikman et al., 2013; Struchen et al., 2008). These studies, in addition to investigations of other essential socioemotional processes such as emotion regulation (Cooper et al., 2015), point to the importance of social communication and emotional interaction in clinical recovery. In a study by Hillis and colleagues, 50% (7/14) of right-hemisphere stroke caregivers reported that “difficulty in understanding the feelings of other people” was in the top five most important consequences of a right-hemisphere stroke (Hillis & Tippett, 2014). It was the most frequently reported problem by right-hemisphere stroke caregivers (above “difficulty in walking”) and was also endorsed by 14% (2/14) of right-hemisphere stroke survivors. But consistent with the lateralization of emotion hypothesis, it was not endorsed by any left-hemisphere stroke caregiver or patient (0/28). Relatedly, we found no evidence that a particular emotional valence (i.e., positive, negative) underpinned emotion recognition deficits in the stroke patients relative to controls. This contradicts the valence hypothesis, which posits that positive emotional information is predominantly processed in the left-hemisphere and negative emotional information is predominantly processed in the right hemisphere (Canli et al., 1998; Davidson, 1995; Wildgruber et al., 2006).

Qualitative reports from right-hemisphere stroke survivors and caregivers have helped establish emotion recognition deficits as a possible culprit for long-term social difficulties (Arntzen et al., 2015; Taule & Råheim, 2014). A loss of fluid social interactions could escalate to a patient’s social withdrawal or exclusion by others, reduced emotional or tangible support, and a reduced capacity to rely on a network of others to stay healthy, happy, and engaged. Our results preliminarily support this by linking emotion recognition difficulties after stroke to two unique aspects of social well-being: withdrawal from social activity and reduced patient-reported social support. Notably, the observed relationship with patients’ loss of social activities could not be explained by changes to general activity levels and was specific to the stroke group. No relationship between loss of social activity and emotion recognition was observed in the control group. This was expected especially because healthy participants rated recent change in social activities to allow for comparison with change since stroke in the patient group. While the control group might show changes in social activity due to extraneous factors such as retirement, these changes would not be expected to correlate with emotion recognition. On the other hand, while only the patient group showed a significant correlation between emotion recognition deficits and reduced social support, a regression model indicated no significant difference in the effect between groups. This result may indicate that emotion recognition, an ability that sharply and consistently declines with age (Ruffman et al., 2008; Schlegel et al., 2014; Schlegel & Scherer, 2016), may be an important factor for maintaining social support networks across healthy and clinical samples in older age. However, caution should be used in interpreting this result due to unexpected group differences in marital status. In addition, we only measured current levels of social support, not recent change, meaning that normal variability unrelated to a stroke could also account for these results.

A few limitations should be considered when interpreting results. Because this was a cross-sectional study, we were unable to fully characterize socioemotional well-being pre-stroke. In the general population, poor social support and isolation have been previously linked to worse health behaviors that increase stroke risk (e.g., reduced physical activity and increased nicotine consumption (Hawkley et al., 2006, 2009; House et al., 1988)). This suggests the possibility that the right-hemisphere stroke group may have had relatively worse social well-being prior to the stroke. However, calculation of prior activity levels shows no group differences for social activities (p=.827), low-physical-demand leisure (p=.542), high-physical-demand leisure (p=.857), or instrumental activities (p=.187). While this implies that stroke patients did not differ from controls in social or physical activity prior to the stroke, the possibility that these retrospective judgements are influenced by a stroke should be considered a potential limitation. For example, right-hemisphere stroke is frequently associated with limited awareness of the disease and related problems (e.g., anosognosia and anosodiaphoria (Gasquoine, 2015)), impaired magnitude estimation (Mennemeier et al., 2005), and memory problems (Gillespie et al., 2006). Because our study measures relied on patient report, these factors should be considered when interpreting results.

An additional limitation is our relatively small sample size, indicating that replication in larger samples is needed. Small sample sizes reduce statistical power and diminish the reliability of multivariate analyses. The multivariate regression models reported here included necessary covariates for this sample, but may have limited generalizability and replication will be essential to confirm results. The trend-level group differences (for positive emotion recognition accuracy and negative emotion recognition accuracy) should also be interpreted with particular caution. While we focused on strokes affecting the MCA-territory, which are the most common, additional investigation in large samples would be needed to identify whether lesions to areas outside of the MCA-territory may also be important for social well-being outcomes. Furthermore, because social networks can include family, friends, co-workers, neighbors and acquaintances, it may be useful for future research to more precisely measure patients’ connections. Recent work points to friendships as being at particular risk for decline after right-hemisphere stroke, particularly in patients with communication deficits (Hewetson et al., 2018). Gaining a better sense of which social contacts are most important and which are most at-risk following stroke could help guide intervention research (Zencius & Wesolowski, 1999).

The GERT-S emotion recognition task does not permit assessment of performance on individual emotions and is relatively cognitively demanding for older adults. Subjects watched each stimulus only one time, then selected from one of fourteen answer choices on a decision screen. Performance accuracy for the control sample was low (49%) although similar to prior studies (Schlegel & Scherer, 2016, Schlegel personal communication), making it difficult to precisely define emotion recognition deficit from these data. Future work in our laboratory adapts the GERT-S task to be easier for older adults and introduces a matched, non-emotional control task to improve interpretability. Lastly, we recognize that an array of interrelated cognitive and socioemotional difficulties may present after stroke (e.g., mood disorders, working memory deficits). In fact, communication deficits caused by any type of cognitive impairment following right-hemisphere stroke have also been linked to reduced social activity engagement in the chronic phase of recovery (Hewetson et al., 2018). However, targeting rehabilitation research on precise domains that: 1) can be easily measured in clinical or non-clinical settings and, 2) already have established treatment options, would be an optimal approach forward. Emotion recognition is a strong candidate domain because it can be trained to improvement in patient populations—as has been shown in traumatic brain injury (Neumann et al., 2015; Radice-Neumann et al., 2009), children with autism (Berggren et al., 2018; Yun et al., 2017), and individuals with depressive symptoms (Jahangard et al., 2012; Penton-Voak et al., 2012). Furthermore, in addition to remediating emotion recognition deficits in patients, promoting training and education for family members and caregivers about emotional communication impairments may be similarly valuable as has been shown in dementia care (Magai et al., 2002).

In sum, our results implicate emotion recognition as a potentially important clinical marker after right-hemisphere stroke that could help predict a patient’s social outcomes. Additional empirical research on the relationship between emotion recognition deficits and risk of social decline will help increase awareness for clinicians, caregivers and stroke survivors, and may uncover new approaches that can benefit social well-being and quality of life for the millions of Americans affected by stroke.

Supplementary Material

Supp 1

Acknowledgements

We thank the Center for Brain Plasticity and Recovery at Georgetown University and MedStar National Rehabilitation Hospital for support.

Funding

This research was supported by the National Institutes of Health under Grant R21HD095273 to ASG; Georgetown University’s Dean’s Toulmin Pilot Award to ASG; and the National Center for Advancing Translational Sciences of the National Institutes of Health under Grants TL1TR001431 to KO and KL2TR001432 to ASG.

Footnotes

Disclosure statement

The authors report no conflicts of interest

Data availability statement

The data that support the findings of this study are openly available in Open Science Framework at https://osf.io/5n96h. Additional raw data is available from the corresponding author, KO, upon request.

References

  1. Adolphs R, Damasio H, Tranel D, Cooper G, & Damasio AR (2000). A role for somatosensory cortices in the visual recognition of emotion as revealed by three-dimensional lesion mapping. Journal of Neuroscience, 20(7), 2683–2690. 10.1523/jneurosci.20-07-02683.2000 [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Adolphs R, Damasio H, Tranel D, & Damasio AR (1996). Cortical systems for the recognition of emotion in facial expressions. Journal of Neuroscience, 16(23), 7678–7687. 10.1523/jneurosci.16-23-07678 [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Arntzen C, Borg T, & Hamran T (2015). Long-term recovery trajectory after stroke: An ongoing negotiation between body, participation and self. Disability and Rehabilitation, 37(18), 1626–1634. 10.3109/09638288.2014.972590 [DOI] [PubMed] [Google Scholar]
  4. Baum CM, & Edwards D (2001). Activity card sort. Washington University at St. Louis. [Google Scholar]
  5. Baum CM, & Edwards D (2008). Activity Card Sort (ACS): Test manual (2nd Ed). Bethesda, MD: AOTA Press. [Google Scholar]
  6. Beckley MN (2006). Community participation following cerebrovascular accident: Impact of the buffering model of social support. American Journal of Occupational Therapy, 60, 129–135. 10.5014/ajot.60.2.129 [DOI] [PubMed] [Google Scholar]
  7. Berggren S, Fletcher-Watson S, Milenkovic N, Marschik PB, Bölte S, & Jonsson U (2018). Emotion recognition training in autism spectrum disorder: A systematic review of challenges related to generalizability. Developmental Neurorehabilitation, 21(3), 141–154. 10.1080/17518423.2017.1305004 [DOI] [PubMed] [Google Scholar]
  8. Boden-Albala B, Litwak E, Elkind MSV, Rundek T, & Sacco RL (2005). Social isolation and outcomes post stroke. Neurology, 64(11), 1888–1892. 10.1212/01.WNL.0000163510.79351.AF [DOI] [PubMed] [Google Scholar]
  9. Borod JC, Cicero BA, Obler LK, Welkowitz J, Erhan HM, Santschi C, Grunwald IS, Agosti RM, & Whalen JR (1998). Right hemisphere emotional perception: evidence across multiple channels. Neuropsychology, 12(3), 446–458. 10.1037/0894-4105.12.3.446 [DOI] [PubMed] [Google Scholar]
  10. Canli T, Desmond JE, Zhao Z, Glover G, & Gabrieli JDE (1998). Hemispheric asymmetry for emotional stimuli detected with fMRI. NeuroReport, 9(14), 3233–3239. 10.1097/00001756-199810050-00019 [DOI] [PubMed] [Google Scholar]
  11. Centers for Disease Control and Prevention. (2019). BRFSS Prevalence & Trends Data. https://www.cdc.gov/brfss/brfssprevalence/
  12. Chan VWK, Chung JCC, & Packer TL (2006). Validity and reliability of the activity card sort-Hong Kong version. OTJR Occupation, Participation and Health, 26(4), 152–158. 10.1177/153944920602600405 [DOI] [Google Scholar]
  13. Chick CF, Rolle C, Trivedi HM, Monuszko K, & Etkin A (2020). Transcranial magnetic stimulation demonstrates a role for the ventrolateral prefrontal cortex in emotion perception. Psychiatry Research, 284, 112515. 10.1016/j.psychres.2019.112515 [DOI] [PubMed] [Google Scholar]
  14. Colantonio A, Kasl SV, Ostfeld AM, & Berkman LF (1993). Psychosocial predictors of stroke outcomes in an elderly population. Journals of Gerontology, 48(5), 261–268. 10.1093/geronj/48.5.S261 [DOI] [PubMed] [Google Scholar]
  15. Cooper CL, Phillips LH, Johnston M, Radlak B, Hamilton S, & McLeod MJ (2014). Links between emotion perception and social participation restriction following stroke. Brain Injury, 28(1), 122–126. 10.3109/02699052.2013.848379 [DOI] [PubMed] [Google Scholar]
  16. Cooper CL, Phillips LH, Johnston M, Whyte M, & Macleod MJ (2015). The role of emotion regulation on social participation following stroke. British Journal of Clinical Psychology, 54(2), 181–199. 10.1111/bjc.12068 [DOI] [PubMed] [Google Scholar]
  17. Davidson RJ (1995). Cerebral asymmetry, emotion and affective style. In Davidson RJ & Hugdahl K (Eds.), Brain Asymmetry (pp. 361–387). The MIT Press. [Google Scholar]
  18. De Winter FL, Zhu Q, Van den Stock J, Nelissen K, Peeters R, de Gelder B, Vanduffel W, & Vandenbulcke M (2015). Lateralization for dynamic facial expressions in human superior temporal sulcus. NeuroImage, 106, 340–352. 10.1016/j.neuroimage.2014.11.020 [DOI] [PubMed] [Google Scholar]
  19. Dupre ME, & Lopes RD (2016). Marital history and survival after stroke. Journal of the American Heart Association, 5(12), e004647. 10.1161/JAHA.116.004647 [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Dzhelyova M, Jacques C, & Rossion B (2017). At a single glance: Fast periodic visual stimulation uncovers the spatio-temporal dynamics of brief facial expression changes in the human brain. Cerebral Cortex, 27(8), 4106–4123. 10.1093/cercor/bhw223 [DOI] [PubMed] [Google Scholar]
  21. Edwards DF, Hahn M, Baum C, & Dromerick AW (2006). The Impact of Mild Stroke on Meaningful Activity and Life Satisfaction. Journal of Stroke and Cerebrovascular Diseases, 15(4), 151–157. 10.1016/j.jstrokecerebrovasdis.2006.04.001 [DOI] [PubMed] [Google Scholar]
  22. Evans RL, Bishop DS, Matlock AL, Stranahan S, Smith GG, & Halar EM (1987). Family interaction and treatment adherence after stroke. Archives of Physical Medicine and Rehabilitation, 68(8), 513–517. [PubMed] [Google Scholar]
  23. Everard KM, Lach HW, Fisher EB, & Baum MC (2000). Relationship of activity and social support to the functional health of older adults. Journals of Gerontology - Series B Psychological Sciences and Social Sciences, 55(4), 208–212. 10.1093/geronb/55.4.S208 [DOI] [PubMed] [Google Scholar]
  24. Fillenbaum GG (2013). Multidimensional functional assessment of older adults: The Duke Older Americans Resources and Services procedures. Psychology Press. [Google Scholar]
  25. Gasquoine PG (2015). Blissfully unaware: Anosognosia and anosodiaphoria after acquired brain injury. Neuropsychological Rehabilitation, 26(2), 261–285. 10.1080/09602011.2015.1011665 [DOI] [PubMed] [Google Scholar]
  26. Gillespie DC, Bowen A, & Foster JK (2006). Memory impairment following right hemisphere stroke: A comparative meta-analytic and narrative review. Clinical Neuropsychologist, 20(1), 59–75. 10.1080/13854040500203308 [DOI] [PubMed] [Google Scholar]
  27. Glass TA, & Maddox GL (1992). The quality and quantity of social support: Stroke recovery as psycho-social transition. Social Science and Medicine, 34(11), 1249–1261. 10.1016/0277-9536(92)90317-J [DOI] [PubMed] [Google Scholar]
  28. Glass TA, Matchar DB, Belyea M, & Feussner JR (1993). Impact of social support on outcome in first stroke. Stroke, 24(1), 64–70. 10.1161/01.STR.24.1.64 [DOI] [PubMed] [Google Scholar]
  29. Gorelick PB, & Ross ED (1987). The aprosodias: Further functional-anatomical evidence for the organisation of affective language in the right hemisphere. Journal of Neurology, Neurosurgery and Psychiatry, 50(5), 553–560. 10.1136/jnnp.50.5.553 [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Hartman-Maeir A, Soroker N, Ring H, Avni N, & Katz N (2007). Activities, participation and satisfaction one-year post stroke. Disability and Rehabilitation, 29(7), 559–566. 10.1080/09638280600924996 [DOI] [PubMed] [Google Scholar]
  31. Hawkley LC, Masi CM, Berry JD, & Cacioppo JT (2006). Loneliness is a unique predictor of age-related differences in systolic blood pressure. Psychology and Aging, 21(1), 152–164. 10.1037/0882-7974.21.1.152 [DOI] [PubMed] [Google Scholar]
  32. Hawkley LC, Thisted RA, & Cacioppo JT (2009). Loneliness Predicts Reduced Physical Activity: Cross-Sectional & Longitudinal Analyses. Health Psychology, 28(3), 354–363. 10.1037/a0014400 [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Heilman KM, Scholes R, & Watson RT (1975). Auditory affective agnosia. Disturbed comprehension of affective speech. Journal of Neurology Neurosurgery and Psychiatry, 38, 69–72. 10.1136/jnnp.38.1.69 [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Hewetson R, Cornwell P, & Shum D (2018). Social participation following right hemisphere stroke: influence of a cognitive-communication disorder. Aphasiology, 32(2), 164–182. 10.1080/02687038.2017.1315045 [DOI] [Google Scholar]
  35. Hillis AE, & Tippett DC (2014). Stroke Recovery: Surprising Influences and Residual Consequences. Advances in Medicine, 2014, 1–10. 10.1155/2014/378263 [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. House JS, Landis KR, & Umberson D (1988). Social relationships and health. Science, 241(4865), 540–545. 10.1126/science.3399889 [DOI] [PubMed] [Google Scholar]
  37. Jahangard L, Haghighi M, Bajoghli H, Ahmadpanah M, Ghaleiha A, Zarrabian MK, & Brand S (2012). Training emotional intelligence improves both emotional intelligence and depressive symptoms in inpatients with borderline personality disorder and depression. International Journal of Psychiatry in Clinical Practice, 16(3), 197–204. 10.3109/13651501.2012.687454 [DOI] [PubMed] [Google Scholar]
  38. Katz N, Karpin H, Lak A, Furman T, & Hartman-Maeir A (2003). Participation in occupational performance: Reliability and validity of the activity card sort. OTJR Occupation, Participation and Health, 23(1), 10–17. 10.1177/153944920302300102 [DOI] [Google Scholar]
  39. Kesler-West ML, Andersen AH, Smith CD, Avison MJ, Davis CE, Kryscio RJ, & Blonder LX (2001). Neural substrates of facial emotion processing using fMRI. Cognitive Brain Research, 11(2), 213–226. 10.1016/S0926-6410(00)00073-2 [DOI] [PubMed] [Google Scholar]
  40. Knox L, & Douglas J (2009). Long-term ability to interpret facial expression after traumatic brain injury and its relation to social integration. Brain and Cognition, 69(2), 442–449. 10.1016/j.bandc.2008.09.009 [DOI] [PubMed] [Google Scholar]
  41. Lehnerer S, Hotter B, Padberg I, Knispel P, Remstedt D, Liebenau A, Grittner U, Wellwood I, & Meisel A (2019). Social work support and unmet social needs in life after stroke: A cross-sectional exploratory study. BMC Neurology, 19(1), 1–10. 10.1186/s12883-019-1451-y [DOI] [PMC free article] [PubMed] [Google Scholar]
  42. Leleu A, Favre E, Yailian A, Fumat H, Klamm J, Amado I, Baudouin JY, Franck N, & Demily C (2019). An implicit and reliable neural measure quantifying impaired visual coding of facial expression: evidence from the 22q11.2 deletion syndrome. Translational Psychiatry, 9, 67. 10.1038/s41398-019-0411-z [DOI] [PMC free article] [PubMed] [Google Scholar]
  43. Lyons KD, Li Z, Tosteson TD, Meehan K, & Ahles TA (2010). Consistency and construct validity of the activity card sort (modified) in measuring activity resumption after stem cell transplantation. American Journal of Occupational Therapy, 64, 562–569. 10.5014/ajot.2010.09033 [DOI] [PMC free article] [PubMed] [Google Scholar]
  44. Magai C, Cohen CI, & Gomberg D (2002). Impact of training dementia caregivers in sensitivity to nonverbal emotion signals. International Psychogeriatrics, 14(1), 25–38. 10.1017/S1041610202008256 [DOI] [PubMed] [Google Scholar]
  45. McDonald S, Flanagan S, Martin I, & Saunders C (2004). The ecological validity of TASIT: A test of social perception. Neuropsychological Rehabilitation, 14(3), 285–302. 10.1080/09602010343000237 [DOI] [Google Scholar]
  46. Mennemeier M, Pierce CA, Chatterjee A, Anderson B, Jewell G, Dowler R, Woods AJ, Glenn T, & Mark VW (2005). Biases in Attentional Orientation and Magnitude Estimation Explain Crossover: Neglect is a Disorder of Both. Journal Cognitive Neuroscience, 17(8), 1194–1211. 10.1162/0898929055002454 [DOI] [PMC free article] [PubMed] [Google Scholar]
  47. National Institute of Neurological Disorders and Stroke. (2011). NIH Stroke Scale. National Institute of Neurological Disorders and Stroke, Dept. of Health and Human Services: Bethesda, MD, USA. [Google Scholar]
  48. Neumann D, Babbage DR, Zupan B, & Willer B (2015). A randomized controlled trial of emotion recognition training after traumatic brain injury. Journal of Head Trauma Rehabilitation, 30(3), E12–E23. 10.1097/HTR.0000000000000054 [DOI] [PubMed] [Google Scholar]
  49. Nijsse B, Spikman JM, Visser-Meily JM, de Kort PL, & van Heugten CM (2019). Social Cognition Impairments in the Long Term Post Stroke. Archives of Physical Medicine and Rehabilitation, 100(7), 1300–1307. 10.1016/j.apmr.2019.01.023 [DOI] [PubMed] [Google Scholar]
  50. Northcott S, Moss B, Harrison K, & Hilari K (2016). A systematic review of the impact of stroke on social support and social networks: Associated factors and patterns of change. Clinical Rehabilitation, 30(8), 811–831. 10.1177/0269215515602136 [DOI] [PubMed] [Google Scholar]
  51. Ovbiagele B, Goldstein LB, Higashida RT, Howard VJ, Johnston SC, Khavjou OA, Lackland DT, Lichtman JH, Mohl S, Sacco RL, Saver JL, & Trogdon JG (2013). Forecasting the future of stroke in the united states: A policy statement from the American heart association and American stroke association. Stroke, 44(8), 2361–2375. 10.1161/STR.0b013e31829734f2 [DOI] [PubMed] [Google Scholar]
  52. Peirce JW (2007). PsychoPy-Psychophysics software in Python. Journal of Neuroscience Methods, 162(1–2), 8–13. 10.1016/j.jneumeth.2006.11.017 [DOI] [PMC free article] [PubMed] [Google Scholar]
  53. Penton-Voak IS, Bate H, Lewis G, & Munafò MR (2012). Effects of emotion perception training on mood in undergraduate students: Randomised controlled trial. British Journal of Psychiatry, 201(1), 71–72. 10.1192/bjp.bp.111.107086 [DOI] [PubMed] [Google Scholar]
  54. Pitcher D (2014). Facial expression recognition takes longer in the posterior superior temporal sulcus than in the occipital face area. Journal of Neuroscience, 34(27), 9173–9177. 10.1523/JNEUROSCI.5038-13 [DOI] [PMC free article] [PubMed] [Google Scholar]
  55. Radice-Neumann D, Zupan B, Tomita M, & Willer B (2009). Training emotional processing in persons with brain injury. Journal of Head Trauma Rehabilitation, 24(5), 313–323. 10.1097/HTR.0b013e3181b09160 [DOI] [PubMed] [Google Scholar]
  56. Ross ED, & Mesulam MM (1979). Dominant Language Functions of the Right Hemisphere?: Prosody and Emotional Gesturing. Archives of Neurology, 36(3), 144–148. 10.1001/archneur.1979.00500390062006 [DOI] [PubMed] [Google Scholar]
  57. Ross ED, & Monnot M (2008). Neurology of affective prosody and its functional-anatomic organization in right hemisphere. Brain and Language, 104(1), 51–74. 10.1016/j.bandl.2007.04.007 [DOI] [PubMed] [Google Scholar]
  58. Ruffman T, Henry JD, Livingstone V, & Phillips LH (2008). A meta-analytic review of emotion recognition and aging: Implications for neuropsychological models of aging. Neuroscience and Biobehavioral Reviews, 32(4), 863–881. 10.1016/j.neubiorev.2008.01.001 [DOI] [PubMed] [Google Scholar]
  59. Schlegel K, Grandjean D, & Scherer KR (2014). Introducing the Geneva Emotion Recognition Test: An example of Rasch-based test development. Psychological Assessment, 26(2), 666–672. 10.1037/a0035246 [DOI] [PubMed] [Google Scholar]
  60. Schlegel K, & Scherer KR (2016). Introducing a short version of the Geneva Emotion Recognition Test (GERT-S): Psychometric properties and construct validation. Behavior Research Methods, 48, 1383–1392. 10.3758/s13428-015-0646-4 [DOI] [PubMed] [Google Scholar]
  61. Schlegel K, Vicaria IM, Isaacowitz DM, & Hall JA (2017). Effectiveness of a short audiovisual emotion recognition training program in adults. Motivation and Emotion, 41, 646–660. 10.1007/s11031-017-9631-9 [DOI] [Google Scholar]
  62. Seydell-Greenwald A, Chambers CE, Ferrara K, & Newport EL (2020). What you say versus how you say it: Comparing sentence comprehension and emotional prosody processing using fMRI. NeuroImage, 209, 116509. 10.1016/j.neuroimage.2019.116509 [DOI] [PMC free article] [PubMed] [Google Scholar]
  63. Sheppard SM, Keator LM, Breining BL, Wright AE, Saxena S, Tippett DC, & Hillis AE (2020). Right hemisphere ventral stream for emotional prosody identification: Evidence from acute stroke. Neurology, 94(10), e1013–e1020. 10.1212/wnl.0000000000008870 [DOI] [PMC free article] [PubMed] [Google Scholar]
  64. Shimoda K, & Robinson RG (1998). The Relationship between Social Impairment and Recovery from Stroke. Psychiatry Interpersonal & Biological Processes, 61(2), 101–111. 10.1080/00332747.1998.11024821 [DOI] [PubMed] [Google Scholar]
  65. Sliwinska MW, & Pitcher D (2018). TMS demonstrates that both right and left superior temporal sulci are important for facial expression recognition. NeuroImage, 183, 394–400. 10.1016/j.neuroimage.2018.08.025 [DOI] [PubMed] [Google Scholar]
  66. Spikman JM, Milders MV, Visser-Keizer AC, Westerhof-Evers HJ, Herben-Dekker M, & van der Naalt J (2013). Deficits in Facial Emotion Recognition Indicate Behavioral Changes and Impaired Self-Awareness after Moderate to Severe Traumatic Brain Injury. PLoS ONE, 8(6), e65581. 10.1371/journal.pone.0065581 [DOI] [PMC free article] [PubMed] [Google Scholar]
  67. Stiekema APM, Nijsse B, de Kort PLM, Spikman JM, Visser-Meily JMA, & van Heugten CM (2019). The relationship between social cognition and participation in the long term after stroke. Neuropsychological Rehabilitation, 1–15. 10.1080/09602011.2019.1692670 [DOI] [PubMed] [Google Scholar]
  68. Streit M, Ioannides AA, Liu L, Wölwer W, Dammers J, Gross J, Gaebel W, & Müller-Gärtner HW (1999). Neurophysiological correlates of the recognition of facial expressions of emotion as revealed by magnetoencephalography. Cognitive Brain Research, 7(4), 481–491. 10.1016/S0926-6410(98)00048-2 [DOI] [PubMed] [Google Scholar]
  69. Struchen MA, Clark AN, Sander AM, Mills MR, Evans G, & Kurtz D (2008). Relation of executive functioning and social communication measures to functional outcomes following traumatic brain injury. NeuroRehabilitation, 23(2), 185–198. 10.3233/nre-2008-23208 [DOI] [PubMed] [Google Scholar]
  70. Taule T, & Råheim M (2014). Life changed existentially: A qualitative study of experiences at 6-8 months after mild stroke. Disability and Rehabilitation, 36(25), 2107–2119. 10.3109/09638288.2014.904448 [DOI] [PubMed] [Google Scholar]
  71. Tucker DM, Watson RT, & Heilman KM (1977). Discrimination and evocation of affectively intoned speech in patients with right parietal disease. Neurology, 27(10), 947–950. 10.1212/wnl.27.10.947 [DOI] [PubMed] [Google Scholar]
  72. U.S. Census Bureau. (2018). American Community Survey: ACS 1-Year Estimates Subject Tables.
  73. Vogt TM, Mullooly JP, Ernst D, Pope CR, & Hollis JF (1992). Social networks as predictors of ischemic heart disease, cancer, stroke and hypertension: Incidence, survival and mortality. Journal of Clinical Epidemiology, 45(6), 659–666. 10.1016/0895-4356(92)90138-D [DOI] [PubMed] [Google Scholar]
  74. Waldron-Perrine B, & Axelrod BN (2012). Determining an appropriate cutting score for indication of impairment on the Montreal Cognitive Assessment. International Journal of Geriatric Psychiatry, 27(11), 1189–1194. 10.1002/gps.3768 [DOI] [PubMed] [Google Scholar]
  75. Weintraub S, Mesulam M -Marsel, & Kramer L. (1981). Disturbances in Prosody: A Right-Hemisphere Contribution to Language. Archives of Neurology, 38(12), 742–744. [DOI] [PubMed] [Google Scholar]
  76. Wellman B, & Wortley S (1990). Different Strokes from Different Folks: Community Ties and Social Support. American Journal of Sociology, 96(3), 558–588. 10.1086/229572 [DOI] [Google Scholar]
  77. Wildgruber D, Ackermann H, Kreifelts B, & Ethofer T (2006). Cerebral processing of linguistic and emotional prosody: fMRI studies. Progress in Brain Research, 156, 249–268. 10.1016/S0079-6123(06)56013-3 [DOI] [PubMed] [Google Scholar]
  78. Witteman J, Van Heuven VJP, & Schiller NO (2012). Hearing feelings: A quantitative meta-analysis on the neuroimaging literature of emotional prosody perception. Neuropsychologia, 50(12), 2752–2763. 10.1016/j.neuropsychologia.2012.07.026 [DOI] [PubMed] [Google Scholar]
  79. Witteman J, Van Ijzendoorn MH, Van de Velde D, Van Heuven VJJP, & Schiller NO (2011). The nature of hemispheric specialization for linguistic and emotional prosodic perception: A meta-analysis of the lesion literature. Neuropsychologia, 49(13), 3722–3738. 10.1016/j.neuropsychologia.2011.09.028 [DOI] [PubMed] [Google Scholar]
  80. Yun SS, Choi JS, Park SK, Bong GY, & Yoo HJ (2017). Social skills training for children with autism spectrum disorder using a robotic behavioral intervention system. Autism Research, 10(7), 1306–1323. 10.1002/aur.1778 [DOI] [PubMed] [Google Scholar]
  81. Yuvaraj R, Murugappan M, Norlinah MI, Sundaraj K, & Khairiyah M (2013). Review of emotion recognition in stroke patients. Dementia and Geriatric Cognitive Disorders, 36(3–4), 179–196. 10.1159/000353440 [DOI] [PubMed] [Google Scholar]
  82. Zencius AH, & Wesolowski MD (1999). Is the social network analysis necessary in the rehabilitation of individuals with head injury? Brain Injury, 13(9), 723–727. 10.1080/026990599121278 [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supp 1
NIHMS1675568-supplement-Supp_1.docx (951.1KB, docx)

RESOURCES