Abstract
This article addresses the use of self-regulatory speech (SRS) in children with specific language impairment (SLI). The authors define the term “self-regulatory speech” as speech that is used for the benefit of one’s self. This includes overt private and social speech that has a self-regulatory function during planning and problem solving, as well as more covert inaudible muttering. Vygotsky (1987) theorized that this “speech for the self” is essential for basic mental operations and executive functions including self-regulation, planning, and problem solving. SRS has been studied in typically developing (TD) children and children with attention deficit hyperactivity disorder (ADHD), but little research has compared children with SLI with TD children. Children with SLI often exhibit executive function (EF) deficits (Henry, Messer, & Nash, 2012; Kapa, Plante, & Doubleday, 2017), so it is likely that they would also exhibit deficits in SRS.
SRS develops during the preschool-early elementary years. TD preschool children often talk overtly through their activities to regulate their thoughts and behaviors (Winsler, Manfra, & Diaz, 2007). They use their language, both social speech directed to others and private speech directed to themselves, to plan and problem solve by talking through the required steps to keep themselves on track (Sturn & Johnson, 1999). Researchers have reported that SRS has a curvilinear developmental trajectory. In the beginning, SRS involves both private and social speech. As children develop, their use of overt private speech increases up to the point where overt, oral speech starts to become internalized, which is initially evident as inaudible muttering before speech becomes completely internal to form silent verbal thought.
SRS increases with age in very young children (e.g., 2- to 3-year-olds, Furrow, 1984; 3.5- to 5-year-olds, Berk & Spuhl, 1995) and declines with age in school-aged children (e.g., 5- to 17-year-olds; Winsler & Naglieri, 2003). In addition to age-related changes, the amount of private speech tends to increase as the task gets more challenging and then to reduce and be ineffective when the task gets too difficult (Duncan & Pratt, 1997). The importance of language for planning and problem solving has been demonstrated using experimental studies. When language use is prevented, cognitive performance during planning and self-regulation is negatively affected. Children with SLI have deficits in expressive language, making it difficult for them to use SRS. In addition, children with SLI often have a receptive language impairment that may limit their comprehension of, interaction with, and verbal scaffolding by people around them. Therefore, children with SLI may have a double barrier to SRS development.
Purposes of This Study
To examine the planning and problem-solving performance (measured by the Tower of London) and SRS of young children with SLI and their TD peers across the first 3 years of school.
To examine the impact of both SLI and hyperactive and inattentive behaviors on SRS and planning and problem solving by comparing the SRS and TOL performance of children with SLI with and without hyperactive and inattentive behaviors with the performance of TD children with and without hyperactive and inattentive behaviors.
To examine the SRS produced on individual items of the TOL and subsequent performance on these items.
The Study
Participants
One hundred seventy-eight 4- to 7-year-old children were recruited from primary schools; 84 were TD children and 94 were children with SLI. Children were required to have nonverbal IQ scores above 85 on the Wechsler Non-Verbal Scale of Ability (WNV), and to pass the practice items on the TOL. In addition, children with SLI had to score below average on the general communication composite (GCC) of the Children’s Communication Checklist–Second Edition (CCC-2), while the TD children were required to have at least average scores on this test.
Assessments
Behavior rating
Teachers completed the Strengths and Difficulties Questionnaire (SDQ; Goodman, 2001) on each child. The researchers in this study were particularly interested in the Hyperactivity and Inattentiveness subscale. High scores (7–10) indicate a substantial risk of clinically significant problems in this area (Goodman, 2001), and children with scores in this range were categorized as having hyperactive and inattentive behaviors. (Note. The SDQ is a screener for ADHD; it is not used for diagnostic purposes).
Planning task
Shallice’s (1982) mechanical TOL test was used to measure planning and problem-solving abilities. The mechanical form of the TOL used a pair of identical wooden game boards, one each for the child and the examiner. The game boards consisted of three pegs: short, medium, and tall, which hold one, two, or three balls of the same size, respectively. Each ball was either yellow, blue, or red. The children were asked to change the configuration of the balls on the pegs of their board to match the ball configuration on the examiner’s board in the minimum number of moves. Problems could be solved in a minimum of three moves up to a maximum of seven moves per item. The examiner concluded the instructions by saying, “Some children like to talk when playing this ball puzzle game, if you want to talk, it is fine” (see Figure 1). All children were required to pass two or more practice items before beginning the test items, starting with the simplest. Each problem proceeded until either the child completed the problem, gave up, or violated a rule. There were four practice items and 17 test problems. Each item was given a score out of 3. A score of 3 was given when a child solved the problem during the first trial in the minimum number of moves. A score of 2 was given when a child solved the problem during the first trial but not in the minimum number of moves. A score of 1 was given when a child solved the problem during second trial of an item. A total scored was derived by summing the scores across all items.
Tower of London task
Coding of speech
Children’s speech during the TOL administration was coded in terms of utterances. An utterance was defined as a length of continuous speech with less than a 2 s gap between words. Mean utterances per item were calculated by averaging the total number of utterances across the total number of TOL items attempted. The children’s utterances were coded for task relevance (planning or non-planning speech), addressee (private or social speech), and inaudible muttering.
Task relevance/irrelevance
Task relevant utterances were all utterances that were directly related to solving the TOL. This could be utterances for formulating a plan, focusing and sustaining attention, evaluation, or seeking assistance. Examples of task relevant utterances are “the blue one goes here” and “I’m looking carefully.” Utterances were coded as task irrelevant if they did not relate to the TOL such as remarks and descriptions of other objects, events, or people outside the testing context such as “I went to the library just now.”
Addressee
All utterances were deemed social speech when the child was interacting with the examiner. This was operationally defined as having any one or a combination of these elements:
eye contact or eye gaze;
using the examiner’s name;
asking a question and waiting for a response from the examiner; or
initiating physical proximity such as touching the examiner’s hand or leaning toward the examiner.
An utterance was coded as private speech after looking for evidence of social speech and finding no evidence of social intent.
Inaudible muttering
This was coded separately from private speech. Inaudible muttering was operationally defined as any whispering that was not loud enough to be understood, muttering, or silent lip movements.
Results
Children with SLI performed significantly lower on the TOL task than the TD children across all ages.
TD children produced more inaudible mutterings than children with SLI.
There was no significant differences in private speech between children with SLI and TD children, nor did amount of private speech differ across the grades.
There were no significant differences between children with SLI and TD children in social speech. There was no significant difference in social speech of the SLI children across the grades, but the TD children produced significantly more social speech in kindergarten than in preprimary and first grade.
Non-hyperactive/non-inattentive children with SLI had higher TOL scores than hyperactive/inattentive children with SLI. Furthermore, children with SLI who were hyperactive/inattentive committed more rule breaks than children with SLI who were not hyperactive/inattentive. TOL scores did not differ for hyperactive/inattentive and non-hyperactive/inattentive TD children.
Children with SLI with hyperactivity produced significantly more private speech than those without hyperactivity. There was no difference in inaudible muttering between these two groups.
More TD children performed better on the TOL without SRS than with SRS; in contrast, more children with SLI performed better when they used SRS than when they did not.
Interpretation
How Do Children With SLI Compare With TD Children on Problem Solving and SRS?
Children with SLI performed more poorly on the TOL than their TD peers. Although performance improved across grade levels for both groups of children, children with SLI were impaired relative to their TD peers at each grade level. This suggests that the planning and problem-solving ability of children with SLI may be lagging behind their TD peers by up to 2 years during the early years of schooling. The fact that such a gap in performance is evident at such a young age is concerning.
An important finding was that TD children exhibited significantly more inaudible muttering compared with children with SLI who had almost no inaudible muttering. This shows that even at a very early age, TD children have inaudible muttering, which is a sign of SRS internalization. They are not only using SRS, but they are also showing signs of internalizing their SRS to form complete verbal thought. In contrast, children with SLI seem to have a delay in inaudible muttering, which may be an indication of an overall delay in SRS development.
Is Planning Performance Associated With Levels of Hyperactivity and Inattention in Both Children With SLI and Their TD Peers?
The results indicated that children with SLI with hyperactive and inattentive behaviors performed more poorly on the TOL than children without hyperactive and inattentive behaviors. The results indicated that the presence of hyperactive and inattentive behaviors further impedes the already impaired planning and problem-solving performance of children with SLI, and that this may be partially explained by the greater number of rule breaks committed by these children during the TOL. These findings suggest that having a SLI and hyperactive and inattentive behaviors may result in a double deficit for these children. This double deficit would be especially problematic because such children would be limited in terms of the compensatory strategies available to them to help them stay on track, plan, and problem solve. Within the SLI group, children with hyperactive and inattentive behaviors produced more private speech than children without hyperactive and inattentive behaviors, but there were no differences in inaudible muttering or social speech.
The SLI children with hyperactivity/inattention may have used more SRS during planning and problem solving because they needed to use SRS to regulate their behavior so that they were able to focus on the task. They were using SRS for two functions: behavioral self-regulation and the cognitive task at hand. Thus, despite having more private speech, their planning and problem-solving performance may still be lower because they need to divide their cognitive resources between these two concurrent functions. Given that 25% to 45% of children with SLI exhibit some degree of hyperactive and inattentive behaviors, speech-language pathologists should consider that these children have a double deficit that impacts their language, SRS, and executive functioning.
Relationship of SRS to Performance on the TOL?
Children with SLI made the most errors on the TOL when they were silent. In contrast, TD children outperformed children with SLI regardless of whether they were talking or silent. Perhaps the TD children who were silent were using silent verbal thought (i.e., not exhibiting signs of SRS as it is fully internalized) while children with SLI who were silent lack verbal thought to support performance on the TOL. The researchers noted that children with SLI displayed virtually no inaudible muttering, which is typically an indication of SRS being internalized. Children with SLI who were silent may not have been using verbal mediation as a strategy during planning and problem-solving tasks compared with children with SLI who have private speech or social speech.
Implications
Because the presence of hyperactivity/inattention with SLI further reduced the performance of children with SLI, the researchers recommend that young children with SLI be screened for inattention and hyperactivity. Because of the combined impact of SLI and hyperactive and inattentive behaviors, an assessment of behavior would provide an important insight that could be used to guide intervention.
Because of the likely deficits in planning and problem solving, the authors recommend that the executive functioning of young children with SLI be assessed. This is particularly important in the early years to enable the provision of appropriate intervention programs for children with SLI. Early intervention and continued monitoring may assist children with SLI develop the planning and problem-solving skills needed for more complex social interactions as they age.
In children with SLI, those who had SRS did better than children who were silent, supporting Vygotsky’s notion that SRS is vital in cognitive tasks. SRS interventions and verbal scaffolding have also proven to be useful for TD children in enhancing their planning and problem-solving abilities (Winsler, 2009). Therefore, SRS training may be useful for children with SLI to assist them in their SRS development.