Background Using Stimulus Pairing Procedures to Up to half of young - PDF document

11/21/2017 Background Using Stimulus Pairing Procedures to Up to half of young children diagnosed with ASD present with language delay (Hudry et al, 2010; McCann et al., 2005) Increase Vocalizations OFFICE OF XYZ Anna Ingeborg Petursdottir Vocal communication is typically among the goals of early Texas Christian University intervention for non‐speaking toddlers and young preschoolers (see e.g., Leaf & McEachin, 1999; Lovaas, 2003; Sundberg & Partington, 1998; Sundberg, 2008) . May be difficult to accomplish for children who do not engage in vocal play or echolalia or produce many recognizable phonemes Background Stimulus pairing (a.k.a. “stimulus‐stimulus pairing”) refers to temporally contiguous presentations of speech sounds and preferred items WHY WOULD WE EXPECT IT TO WORK? Recommended in some early intervention texts as a way to increase child vocalizations (Greer & Ross, 2008; Sundberg & Partington, 1998; Sundberg, 2007) Common questions: • Why would we expect this to work? • Does it work? • How should I do it? 1

11/21/2017 Canonical Syllables Cooing Variability, prosody, stimulus control Why do Birds Sing? Social feedback (e.g., King & West, 1983; King, West, & Goldstein, 2005; Smith, King, & West, 2000; Marler & Nelson, 1993; West & King, 1988) 2

11/21/2017 Why do Babies Babble? Why do Babies Babble? Parents respond socially to a majority of infant prelinguistic Contingent vocalizations (e.g., Gros‐Louis, West, Goldstein, & King, 2006) attention increases rate and phonological quality of vocalizations (Goldstein, West, & King, 2003; Wu & Gros‐ Louis, 2016) Why do Babies Babble? Why do Babies Babble? But... infants babble more when playing The greater the density of contingent attention, the sooner the alone than when interacting with adults child achieves early language milestones (Tamis‐Lemonda, Bornstein, & (Harold & Barlow, 2013) Baumwell, 2001) 3

11/21/2017 Why do Babies Babble? Why do Babies Babble? Native If parents respond language to all infant and prosody vocalizations with of crying specific types of (Mampe et al., vocalizations of 2009; Wermke their own, the et al., 2016) infants’ production of similar vocalizations increases (Goldstein & Schwade, 2008) Automatic Reinforcement Why do Birds Sing? Long been hypothesized to operate in the Social feedback (e.g., King & acquisition of human babbling (e.g., Mowrer, West, 1983; King, West, & Goldstein, 2005; 1960; Skinner, 1957) Smith, King, & West, 2000; Marler & Nelson, 1993; West & King, 1988) Auditory feedback from own song after exposure to mature song (see Brainard & Doupe, 2000) 4

11/21/2017 How can we establish speech sounds How can we establish speech sounds if a child is not making any? if a child is not making any? Socially mediated Automatic Socially mediated Automatic reinforcement: reinforcement: reinforcement: reinforcement: Supplement social Increase the Supplement social Increase the attention with more reinforcing value of attention with more reinforcing value of effective reinforcers speech sounds effective reinforcers speech sounds Shaping Speech Sounds Via Contingent Using Lag Reinforcement to Increase Delivery of Preferred Items Variability of Vocalizations Demonstrated in a number of studies (e.g., Drash, High & Tudor, 1999; Fineman, 1968; Lovaas, Berberich, Perloff, & Schaeffer, 1966) Drawbacks: • Requires a therapist skill Requires phoneme variability in baseline that is difficult to teach • Time‐consuming Implementation requires fine discriminations between • Not always possible phonemes (Lovaas, 2003) 5

11/21/2017 How can we establish speech sounds if a child is not making any? Socially mediated Automatic reinforcement: reinforcement: Supplement social Increase the attention with more reinforcing value of effective reinforcers speech sounds Stimulus Pairing Procedures “Bah” DOES IT WORK? 6

11/21/2017 Stimulus Pairing Procedures Stimulus Pairing Procedures Reliable effect on child vocalizations demonstrated in Failures to produce the effect have also been reported several studies (Esch, Carr, & Grow, 2009; Lepper & Petursdottir, 2017; in the literature (Carroll & Klatt, 2008; Esch, Carr, & Michael, 2005; Miguel et Lepper, Petursdottir, & Esch, 2013; Miguel, Carr, & Michael, 2002; Miliotis et al., 2012; al., 2002; Normand & Knoll, 2006; Stock, Schulze, & Mirenda, 2008; Yoon & Feliciano, Rader et al., 2014; Smith et al., 1996; Sundberg et al., 1996; Yoon & Bennett, 2000) 2007) • The resulting vocalizations are not echoic and not a product of • Novel intervention and “implausible” effect – easier to incidental reinforcement by the preferred items publish negative results? • The effect is not due to repeated modeling of sounds or • Procedural variability across studies presentation of preferred items; pairing is necessary • Little is known about variables that impact outcomes • Paired items must be reinforcers (Smith et al., 1996) But does it really work? Although many studies have demonstrated effects on speech‐like vocalizations, that does not mean we have demonstrated either the efficacy or the effectiveness of the procedure Moderate overall effect size Does the increase in speech sounds actually help establish functional vocal communication? Carroll and Klatt (2008): • Effect of stimulus pairing on the vocalizations of 1 of 2 Positive outcomes for participants • That participant’s target vocalization was successfully two‐thirds of all brought under echoic control participants 7

11/21/2017 But does it really work? But does it really work? A few studies have additionally demonstrated that A few studies have additionally demonstrated that after vocalizations emerge via stimulus pairing, after vocalizations emerge via stimulus pairing, direct differential reinforcement can produce a further rate reinforcement can produce a further rate increase (Esch et al., 2009; Lepper & Petursdottir, 2017; Yoon & Feliciano, 2007) increase (Esch et al., 2009; Lepper & Petursdottir, 2017; Yoon & Feliciano, 2007) But no assessment of mand function reported But no assessment of mand function or other verbal operant function reported Lepper & Petursdottir (2017): Experiments 1 and 2 Lepper & Petursdottir (2017): Experiments 1 and 2 Experiment 3; not included in published version Experiment 3; not included in published version 0.8 0.8 Gabe Gabe 0.7 0.7 Participant Gabe’s Participant Gabe’s 0.6 0.6 history: history: Contingent Contingent 0.5 0.5 Free Access Free Access 0.4 0.4 (1) Two sounds (“day” (1) Two sounds (“day” 0.3 0.3 and “moo”) paired and “moo”) paired 0.2 0.2 with the same with the same Vocalizations per Min Vocalizations per Min 0.1 0.1 preferred item. Rate preferred item. Rate Day Day 0 0 of emission of both of emission of both 0 1 2 3 4 5 6 7 8 0 1 2 3 4 5 6 7 8 0.8 0.8 sounds increased. sounds increased. 3 Min Sessions 3 Min Sessions 0.7 0.7 0.6 0.6 (2) Differential (2) Differential 0.5 0.5 reinforcement of reinforcement of 0.4 0.4 both sounds with both sounds with 0.3 0.3 different preferred different preferred 0.2 0.2 items; rate items; rate 0.1 0.1 Moo Moo increased further increased further 0 0 1 2 3 4 5 6 7 8 1 2 3 4 5 6 7 8 Sessions Sessions 8