Excerpted from chapter 1 of DO-WATCH-LISTEN-SAY: Social and Communication Intervention for Children with Autism, by Kathleen Ann Quill, Ed.D.
Copyright © 2000 by Paul H. Brookes Publishing Co. All rights reserved.
The Complexity of Autism
Autism is characterized by a triad of impairments in the areas of socialization, communication, and ritualistic behavior (American Psychiatric Association, 1994). Within the spectrum of autism disorders, there are different subgroups with various developmental profiles (Bristol et al., 1996; Szatmari, 1992; Wing & Attwood, 1987). Because autism is not a singular disorder, it is unlikely that one intervention approach will benefit all subgroups equally. In order to develop a range of intervention strategies, an understanding of this complex developmental disorder is necessary.
The following cases of children diagnosed with autism highlight the diversity of the disorder. These examples show how socialization, communication, and ritualistic behavior vary significantly among children with autism.
Paul is an anxious 4-year-old child who has many fears and phobias. His language development has followed normal milestones in terms of vocabulary and sentence complexity. Since toddlerhood, he has been fascinated with letters of the alphabet and numbers. He runs to the television for all commercials. He collects and carries toys — always blue — in sets of three. He enjoys solitary play and has always been adept with constructive toys and puzzles. He observes peers playing but rarely interacts with them. Paul's parents report that it is difficult to engage their son in various topics of conversation. He only wants to talk about the alphabet, numbers, and his many fears, and he does so incessantly.
Matty is 4 years old and a very active child. His language is delayed and he currently speaks in one- or two-word phrases, although he can recite segments from his favorite Disney videos. His play consists of lining up cars, trains, and blocks. He enjoys books and computers. He has frequent tantrums, particularly when his activity is interrupted or there is a change in his routine. He shows no interest in peers.
Andrew is 4 years old and very self-absorbed. He is nonverbal, and his only vocalizations are protests and crying. He pulls and pushes adults to make basic requests, or he independently gets what he needs. He occupies his time by playing with ribbons, string, and any object that twirls. He has difficulty sleeping through the night and is a very picky eater. He runs away and covers his ears in the presence of other children.
It is clear that each child displays the social, communication, and ritualistic behaviors associated with autism, although in quite diverse ways. The severity of the social and communicative impairments and ritualistic behaviors varies from mild to severe. For example, Paul has mild social and communication difficulties and severe ritualistic behaviors. Matty displays relatively moderate impairments in social development, communication development, and ritualistic behaviors, while Andrew presents with severe impairments in all three areas. The profiles of these three children also reflect the diversity of the cognitive and language impairments that accompany autism.
Autism commonly occurs in association with other developmental disabilities, including a wide range of related motor, language, and cognitive impairments. Motor impairments range from motor planning difficulties to severe dyspraxia (Hanschu, 1998). Language impairments vary from mutism to atypical language acquisition or language regression (Prizant, 1996). Many children with autism also have some degree of mental retardation (Lord, 1996). Each of these developmental differences contributes to the tremendous variability observed in autism.
THE NATURE OF AUTISM
Research into the origins of the developmental disabilities that accompany autism has shown that the cognitive, language, communication, and social differences interface in complex ways. There is a growing understanding that an inability to process and understand social and affective information in a cohesive manner may lie at the core of autism (Baron-Cohen, 1995; Frith, 1989; Hobson, 1996).
According to Frith (1989), a cluster of cognitive impairments constrains the child's ability to integrate information in a cohesive and flexible manner. A child with autism is less able to process multiple language, social, and emotional messages simultaneously. This limits the ability to organize the dynamic and unpredictable quality of typical social interaction, fragments understanding of social-communicative experiences, and contributes to rigid, repetitive, and atypical social-communicative interactions.
According to Baron-Cohen (1995), the core impairment is in processing social information. The capacity to understand the thoughts and feelings of others, termed theory of mind, is believed to be profoundly impaired in children with autism. Difficulties with social perspective-taking affect the ability to understand, predict, and respond to the social, communication, and emotional behaviors of others.
According to Hobson (1996), autism is an inability to perceive and understand emotional expressions. This limits the capacity to share attention and emotion with others. It also explains poor social attention, limited social initiations, and difficulty with social reciprocity.
Although the research views each impairment separately, it is more likely that the inability to interpret emotion, understand social perspectives, and integrate information are interrelated. These impairments underlie the social and communicative deficits in autism. The complexity and relationship of these impairments must be taken into account when developing interventions.
COGNITION IN AUTISM
Attention, information processing, and social-cognition can be impaired in children with autism. In general, difficulties with the integration of information, abstract reasoning, and cognitive flexibility have been reported (Minshew, Goldstein, Muenz, & Payton, 1992). Table 1.1 summarizes the cognitive profile observed in autism. As indicated in the table, differences have been found in attentional processes, processing strategies, and social-cognition. Each cognitive weakness lies on a continuum. The severity can be defined as the degree to which each of these cognitive processes dominates thinking and behavior.
In order to attend to others, it is necessary to feel comfortable, filter out distractions, and know what is relevant. Research has indicated that the majority of children with autism show hypersensitivities or atypical responses to sensory stimulation (O'Neill & Jones, 1997; Ornitz, 1989). Adults with autism often describe experiences of feeling overwhelmed (Barron & Barron, 1992; Grandin, 1995a; Williams, 1992). Problems with sensory processing cause distractibility, disorganization, and discomfort, and the result is a tendency to hyperfocus or be drawn to the same repetitive stimulus. Repetition is viewed, in part, as a means to create order amid chaos.
Additional research has suggested that children with autism exhibit overselectivity, meaning that it is difficult for them to attend to the multiple features inherent in all stimuli (Lovaas, Koegel, & Schreibman, 1979). They also have difficulty determining the most meaningful feature of a given stimulus (Frith & Baron-Cohen, 1987). They focus on fewer features or less relevant cues. As the number of cues in natural contexts increases, responses to those cues decrease (Pierce, Glad, & Schreibman, 1997). Given that learning requires attention to multiple environmental features, attentional difficulties markedly impede development, particularly in the social domain. For example, in order to understand the meaning of someone's message, it is necessary to attend simultaneously to the speaker's words, facial expression, tone of voice, and body gestures as well as the social context. If the child is preoccupied or attending to only one cue, such as someone's gestures, he will be less likely to understand the full meaning of the social message.
Other studies of attentional processes report that children with autism have difficulty shifting attention between visual and auditory stimuli (Ciesielski, Courchesne, & Elmasian, 1990; Courchesne, 1991). This impairs their ability to follow the rapid pace and complex features of social interaction. Children struggle to attend to rapidly occurring sensory, language, and social events.
Cognitive processing is a multilayered system in which information is analyzed, organized, stored, and remembered in complex ways. A person instantly links a new experience to a number of related past experiences. The novel information reshapes and redefines all related knowledge. For instance, a typically developing child sees a new animal. She notices all of its separate attributes through different sensory channels, listens to language information about the animal, and associates certain emotions with the animal. She links all of the separate attributes to similar features in other animals (real and imaginary), assigns meaning to her sensory perceptions, combines all of the pieces of language information to her existing body of knowledge about animals, and creates a new and revised concept of animals. In a few moments, she has a new meaning about animals, and the information is stored in an infinite number of ways. This happens unconsciously and effortlessly. Most important, because this animal concept is multilayered and connected to related experiences, there can be an infinite number of ways to recall the new experience.
The way children with autism process information sharply contrasts with the complex and fluid style of a typically developing child. Problems integrating information in meaningful and flexible ways characterize their processing style. First, overselective attention results in a tendency to process information one piece at a time. Thus, a narrow and more restricted understanding of information emerges. Second, information that is fixed in space is more easily processed than rapidly occurring events. Third, material is often stored and remembered as a whole rather than reorganized and integrated in a flexible manner. These processing constraints restrict concept development, generalization, and social-communicative interaction. For example, social interaction requires the simultaneous integration of many sensory, language, social, and affective cues. It demands flexible attention, an understanding of what is relevant, and the ability to disregard what is not pertinent.
For decades, autism has been described as a problem with understanding the meaning of experiences (Wing, 1988). Wing considered the challenges of autism to reflect impairments in one's ability to
Assign meaning to perceptions in a multitude of ways
Infer meaning by moving beyond what is seen or heard
Infer meaning from the social cues of others
Extract meaning from the emotional cues of others
Children with autism attend to specific, observable information and are less able to look beyond pure perception to assign meaning to their experiences. A propensity toward narrow and restricted conceptual understanding characterizes this learning pattern. Children with autism tend to make illogical concrete associations, especially with regard to social and emotional experiences.
Learning is an active process whereby experiences are assimilated into complex networks of meaning. In the absence of understanding the meaning of an experience, information is merely stored as a whole, unrelated to anything else. For example, when one memorizes the Pledge of Allegiance, it is probably stored in the "American flag file" without really understanding its specific meaning. Gestalt processing, when information is processed and remembered as a whole, generally reflects a decreased understanding of the meaning of the parts.
When children with autism are unable to extract meaning from information in a flexible, integrated way, they are left with a series of fragmented experiences that they use in a gestalt manner (Prizant, 1982). Gestalt processing is reflected in specific learning and behavior patterns. Echolalic speech, or repeating what others say, and routinized conversations typify gestalt processing.
Research has found that visuospatial information is more easily processed than auditory and transient information in children with autism (Hermelin & O'Connor, 1970). The amount of time the information remains fixed in space influences the child's ability to process. Visuospatial information can be attended to as long as needed for it to make sense, while other inputs, particularly auditory ones, must be encoded instantly. Studies have indicated that children with autism perform best on intelligence test tasks such as matching, block design, object assembly, and pattern analysis (DeMyer, 1975; Harris, Handleman, & Burton, 1990; Lincoln, Courchesne, Kilman, Elmasian, & Allen, 1988; Siegel, Minshew, & Goldstein, 1996), all of which involve stimuli that remain visible at all times. In her personal account of living with autism, Grandin (1995b) termed this visual thinking, and she emphasized the need to rely on visual images for understanding.
Successful social and communicative interactions require one to quickly attend to and understand the meaning of fleeting visual and auditory information. The transient nature of language and nonverbal social information is more difficult for children with autism to follow and, therefore, contributes to their social and communicative impairments. They struggle to process the rapidly changing social events that are inherent in social interaction.
Finally, research into the memory capabilities of children with autism has further highlighted information-processing differences (Boucher, 1981; Boucher & Lewis, 1988; Boucher & Warrington, 1976; Prior, 1979; Sigman, Ungerer, Mundy, & Sherman, 1987). Memory is a complex process that reflects how information has been organized and stored for use. Memory tasks vary from short term (e.g., immediate repetition) to long term (e.g., accessing from all previously stored information), and it can require recognition (e.g., multiple choice) or recall (e.g., accessing information without explicit cues). Rote memory and recognition tasks remain intact in individuals with autism (Boucher, 1981; Boucher & Lewis, 1988; Boucher & Warrington, 1976; Prior, 1979; Sigman et al., 1987). Rote memory does not require flexible integration of information, and recognition tasks provide explicit retrieval cues. Because children with autism store information in a more restricted manner, their reliance on these patterns can be expected. Memory tasks that require free recall without explicit retrieval cues are more problematic. These researchers suggest that impaired recall memory contributes to the social-communicative deficit. Individuals with autism are less able to gain access to information that is relevant to the rapidly changing social context. They rely more on concrete retrieval cues to remember language information (Tager-Flusberg & Anderson, 1991) and to initiate spontaneous communication.
Typically developing children have a predisposed sensitivity to the feelings of others. An understanding of these feelings results from being able to link emotions and behaviors (Leslie & Frith, 1988). Typically developing infants begin this process of social referencing when they look for a person's reaction to objects and social events (Bruner, 1981). Through social referencing, infants try to understand the socioemotional significance of each experience. Over time, young children develop theory of mind and understand that others have intentions, thoughts, desires, and feelings that differ from their own. This ability to infer the mental states of others allows children to anticipate, comprehend, and predict the social behaviors of others (i.e., "I know what you know"). Most important, this understanding plays a significant role in social-communicative interactions. For instance, in conversation, the child needs to continually monitor what the partner knows and expects in order to make information relevant. Similarly, the child interprets meaning and intent from the verbal and nonverbal behaviors of the partner. Only with this social knowledge can children regulate and adjust their own language and social behavior in social-communicative interactions.
Numerous studies have investigated the degree to which children with autism have difficulty attributing mental states — such as intents, thoughts, and feelings — to themselves and others in order to understand social behavior. Particular deficits in the development of theory of mind have been identified in children with autism. This line of study has prompted additional insights into the core social and communication impairments of autism (see Baron-Cohen, 1995, and Baron-Cohen, Tager-Flusberg, & Cohen, 1993, for a review).
Baron-Cohen, Leslie, and Frith (1985) conducted a seminal study that suggested a specific theory of mind deficit in autism. They tested children's ability to understand different types of social events using a multipicture sequencing task. Three types of sequenced social stories were presented to the children: physical-causal (e.g., a child smiling while swinging on a swing), social-behavioral (e.g., a child crying after someone took his ice cream out of his hand), and mental state (e.g., a child looking confused after someone took his toy when he was not looking). They compared the performance of children with mental retardation and typically developing young children with children of a higher mental age who had autism. The results were striking. In contrast to the children in the control groups, who had no problems with the various social story sequences, the children with autism understood the physical-causal and social-behavioral sequences but had significant difficulty understanding the mental state sequences. Even though all of the stories entailed similar social and emotional content, the children with autism only struggled with social events that required them to take into account what someone else knows or expects in order to predict his or her behavior.
These findings have been replicated — using a variety of tasks, materials, and social events — with similar outcomes (Leslie & Frith, 1988; Perner, Frith, Leslie, & Leekam, 1989). Typically developing 4-year-old children correctly perform tasks that test theory of mind, but most children with autism with a verbal age of 8 years have difficulty with the same tasks. In addition, even when children with autism have been successfully taught theory of mind tasks, they fail to demonstrate an understanding of the concept. They are unable to generalize the concept to new tasks presented in a different format (Ozonoff & Miller, 1995). Furthermore, no measurable effect on the children's ability to develop conversational skills or use mental state terms in their speech has been found (Hadwin, Baron-Cohen, Howlin, & Hill, 1997).
Questions still remain as to whether the theory of mind deficit is a specific impairment of social-cognition. Some research suggests that this phenomenon in autism is actually a social-cognitive delay that correlates with language deficits (Perner et al., 1989) or a result of specific processing demands that impair social problem solving (Pierce et al., 1997).
Perner and colleagues (1989) demonstrated a relationship between language abilities and the acquisition of theory of mind in autism. They found that 17% of the children with autism studied were able to complete theory of mind tasks. Their success with the social-cognitive tasks correlated with higher skills in other areas of verbal cognition. Other studies showed that the children who understood theory of mind tasks demonstrated more flexible conversational skills (Eisenmajer & Prior, 1991) and more social insight (Frith, Happe, & Siddons, 1994).
Pierce, Glad, and Schriebman (1997) found that children with autism were able to answer questions regarding mental states when the task was simplified in terms of processing requirements. They found that the number of social cues influenced the children's ability to interpret social situations. In this study, children were shown videotaped social scenes for which the number of cues leading to the correct understanding of the story varied. Each short scene depicted a social interaction between two children in which an observable action occurred, such as giving a gift. The four types of social cues presented in the scene were verbal (e.g., "I like your toy"), tone of voice (e.g., speaking in an animated voice), nonverbal (e.g., smiling at a person), and nonverbal with an object (e.g., giving someone a present). Children were asked a series of questions to assess their understanding of the social situation. When compared with control groups, children with autism performed equally well on questions relating to stories containing one cue. When stories contained multiple social cues for the correct interpretation of the story, however, the children with autism performed more poorly than those in the control groups. The authors concluded that the complexity of the social environment and the number of social cues that children must simultaneously attend to strongly influence their ability to interpret the meaning of a social situation. When children with autism must simultaneously process multiple social cues within a rapidly changing environment, the attentional and processing demands are too challenging.
Finally, there is a growing recognition of impaired higher order cognitive abilities in autism. The ability to modulate continuous activity, shift problem-solving strategies, and adapt to ongoing information is problematic for children with autism in both social and nonsocial tasks (Ozonoff, 1995; Ozonoff, Pennington, & Rogers, 1991). All of the attentional, processing, and conceptual difficulties reviewed thus far either result in or are influenced by impaired higher order problem solving. Problem-solving flexibility, not IQ score, has been shown to correlate with better social understanding. Similarly, poor social understanding correlates with difficulties in flexible problem solving (Ozonoff, 1995; Russell, 1993). Higher order cognitive strategies allow for integration, inference, analysis, synthesis, creativity, negotiation, assumptions, predictions, anticipation, and clarification of information. It is precisely these skills that are so profoundly affected in the learning, social, and communication behaviors of children with autism.