Decoding actions and emotions in deaf children: Evidence from a biological motion task

This study aimed to explore the recognition of emotional and non-emotional biological movements in children with severe and profound deafness. 24 deaf children, together with 24 control children matched on mental age and 24 control children matched on chronological age, were asked to identify a person’s actions, subjective states, emotions, and objects conveyed by moving point-light displays. Results showed that when observing point light displays, deaf children showed impairments across all conditions (emotions, actions and moving objects) compared to their chronological age matched controls but showed no differences across subjective states. The results are supportive that deaf children present developmental delays in their biological motion apart from the ones relative to their own mental state, and that this may be interpreted in relation to the expertise they have acquired in decoding action toward themselves. The findings are discussed in relation to deaf children viewing motion stimuli very differently to hearing children (e.g. Bosworth & Dobkins, 2002). Deficits in emotional and social interactions have been widely reported in deaf children (Kusché, Garfield, & Greenberg, 1983; Vernon & Greenberg, 1999; Wauters & Knoors, 2007, Weisel & Bar-Lev, 1992). Since the development of linguistic skills and socio-emotional competences have been closely linked (e.g., Malle, 2002), these deficits have been consistently attributed to delays in language acquisition and/or reduced opportunities to converse about personal experiences with other people (e.g., Peterson & Siegal, 1995; 1998). Among the social impairments documented in deaf populations are problems in understanding other people’s emotions (e.g., Rieffe & Terwogt, 2000). These difficulties have often been interpreted as Theory of Mind (ToM) deficits, thereby implicating a lack of awareness of other people’s mental states. Deficits in ToM abilities have been most commonly reported in congenitally deaf children of hearing parents who have not been exposed to sign language (Peterson & Siegal, 1995, 1998, Remmel & Peters, 2008, Russell et al., 1998, Steeds, Rowe & Dowker, 1997) suggesting an early and specific need for interactions with a native speaker. Indeed deficits in ToM abilities have also been observed in late-signing deaf children (Peterson & Slaughter, 2006). Given that the ability to understand emotions is related to measures of social competence (Custrini & Feldman, 1989), ratings of peer popularity and likeability (Denham, McKinley, Couchoud, & Holt, 1990), and academic achievement scores (Nowicki & Duke, 1992), questions about the ways that deaf children access emotion information are important. Experimental studies into social abilities in deaf populations have emphasized delays in understanding other people’s emotions (e.g., Dyck, Farrugia, Shochet & Holmes-Brown, 2004). However to date, studies testing facially expressed emotions in the deaf have produced equivocal findings. Early studies indicated that deaf children do make more errors in recognising facial expressions of emotion than hearing children and deficits are greater in those with pre-lingual than post-lingual hearing loss (Bachara, Raphael & Phelan, 1980; Schiff, 1973). However, in later studies where deaf and hearing children were presented with simple emotion recognition tasks that involved emotion matching, group differences did not emerge (Hosie, Gray, Russell, Scott & Hunter, 1998). It is noteworthy that the majority of research studies that have addressed questions about emotion recognition abilities in deaf populations have relied on photographic, thus static, representations of facial expressions (e.g., Dyck et al., 2004). However dynamic faces have been argued to give a more accurate measure of performance as it is similar to what we experience in everyday life whereas static faces may underestimate a person’s ability to recognise emotions (Moore, 2001). The motion in dynamic faces provides independent information about emotional expression that is not available in posed static faces (Hill & Johnston, 2001) and research has shown that recognition and discrimination of emotions is superior in dynamic compared to static displays (e.g., Bould & Morris, 2008; Werhle et al., 2000). Whilst it is clear that expressions of emotion are powerfully conveyed by faces, movement of body parts are also important as both context and gestures provide vital sources from which we derive emotional meanings (e.g., Clarke, Bradshaw, Field, Hampson, & Rose, 2005; Pollick, Hill, Calder, & Paterson, 2003). For example, when point-lights are attached to the joints of an invisible moving human in a dark room, the visual system can rapidly and reliably distinguish these from similar motion patterns that do not emanate from human beings (e.g., Johansson, 1973). Indeed, a number of studies have shown that complex information, such as emotions, desires, intentions, and dispositions, expressed by a single person or mutual agents can be reliably conveyed using point light cues (Blake & Shiffrar, 2007; Chouchourelou, Toshihiko, Harber & Shiffrar, 2006; Clarke, Bradshaw, Field, Hampson & Rose, 2005). Whilst sensitivity to point-light displays substantially increases during the first 5 years of life Biological Interpretation in Deaf 2 (Pavlova, Kragelch-Mann, Sokolov & Birbaumer, 2001), evidence suggests that such abilities are visible at the earliest stages in perceptual development (e.g., Bertenthal & Pinto, 1994) and may be implicated in the development of non-verbal social communication skills (Dittrich, Troscianko, Lea, Morgan, 1996). For individuals who are deaf or hard of hearing, recognition of emotional expression from body gesture would appear to be vital. However, relatively little is known about the extent to which deaf children utilize such cues. One recent strand of research has provided evidence that deaf children adopt alternative perceptual strategies to hearing children. For example, they rely more heavily on motion cues whilst gauging when to cross a busy road (Bosworth & Dobkins, 2002) and when comprehending critical linguistic information in the hand movements of British sign language (Corina et al., 2007). It is surprising, given both the importance of motion for emotion understanding, and findings suggesting atypical motion processing in deaf children, that motion processing in communication domains has not been widely studied in this population. Evidence supporting the benefit of using motion cues to improve understanding of socio-emotional content has been investigated in other clinical populations, including autism spectrum disorders (ASD). This work has built on earlier findings showing deficits in ToM and emotional face-processing (e.g., Celani, Battacchi & Arcidiacono, 1999). The study, carried out by Parron et al. (2008) investigated action and emotion interpretation using point light displays (PLD) in children with ASD. These findings revealed a dissociation with impaired performance relative to age and intelligence matched controls on PLDs with emotional value, but no impairment relative to controls, on PLDs depicting simple personal actions, subjective states and objects. These findings were consistent with previous studies showing a selective impairment in interpreting PLDs with emotional content in autism (Hubert et al., 2007; Moore, Hobson & Lee, 1997). As previous studies into emotion recognition in deaf children have largely utilised static representations of faces, the nature of the emotion processing deficit is currently unclear. In order to capitalize on their proven strengths in motion processing (Bosworth & Dobkins, 2002; Corina et al., 2007), we adopted the methods used by Parron et al. (2008) and used PLDs to test the assumption of a global emotion-processing deficit in deaf children. Point lights techniques have been used previously with deaf populations to assess how sign language influences action processing (Knapp, Cho & Corina, 2008). In Knapp et al., study, adult deaf signers and hearing non-signers were asked to detect signed and pantomimic point-light movements that were embedded in a field of white noise dots. Whilst the hearing non-signers found it harder to detect differences in sign movements than in pantomimic movements, deaf subjects showed a similar pattern of performance across the two conditions. This difference suggests that deaf participants—compared to hearing participants—do not differentiate signs from other nonlinguistic gestures. In the current study, children with severe or profound hearing difficulties, and hearing controls were tested on their ability to recognize a person’s actions, subjective states, emotions, and objects conveyed by moving point-light displays. The study aimed to ascertain whether children with severe or profound hearing loss recognize basic human actions represented by PLDs and to determine whether discrimination performance is influenced by the emotional content of the displays.