About this paper symposium

---

Panel information
Panel 1. Attention, Learning, Memory

Paper #1
Preschool Children’s Talk Experiences in the Classroom: Variability and Predicto
Author information	Role
Logan Pelfrey, The Ohio State University, United States	Presenting author
Kelly M. Purtell, The Crane Center for Early Childhood Research and Policy at The Ohio State University, USA	Non-presenting author
Laura M. Justice, The Crane Center for Early Childhood Research and Policy at The Ohio State University, USA	Non-presenting author
Tiffany Foster, The Crane Center for Early Childhood Research and Policy at The Ohio State University, USA	Non-presenting author
Dwight Irvin, College of Education, University of Florida, USA	Non-presenting author
Hugo Gonzalez Villasanti, College of Mechanical Engineering, University of Michigan, USA	Non-presenting author
Abstract
Preschool children’s incoming and outgoing talk with teachers and peers in the classroom significantly influences their development of early social skills and vocabulary (e.g., DeLay et al., 2016; Justice et al., 2018). Evidence suggests specific child- and classroom-level factors, such as children’s language and social skills and structural features of the classroom social network, are associated with the teacher and peer- talk children experience (e.g., Chen et al., 2020; Jiang et al., 2023). Nonetheless, our understanding of the nature of individual differences in preschool children’ talk experiences in the classroom remains relatively limited, especially in terms of within and between classroom variability in children’s exposure to talk over the year. This gap is partly attributable to the in-person observation methods used to measure children’s talk within these environments. In-person observations are costly in terms of time and resources, therefore often featuring brief, infrequent observations on which generalizations are made. Additionally, with in-person observation methods, only one child at a time can typically be observed. Sensing systems offer a potential means to overcome these limitations. Sensing systems typically comprise voice recordings, location tracking, and speech processing algorithms and can provide continuous talk and proximity data on all participating children in the classroom simultaneously (Foster et al., 2024). One such system is Interaction Detection in Early Academic Settings (IDEAS; Gonzalez-Villasanti, 2020), which processes data on individual children in a classroom to document their proximity and talk. The present study uses IDEAS to address two aims. First, we seek to determine the extent to which there is within- and between-classroom variability in children’s incoming and outgoing talk and conversational turns with teachers and peers. Second, we seek to identify child- and classroom-level factors associated with children’s incoming and outgoing talk and conversational turns with teachers and peers. We used IDEAS to objectively capture these variables at eight timepoints collected using day-long naturalistic recordings over one academic year. The study sample comprises 21 preschool classrooms/teachers and 136 children (76% white; 52% boys). To address the first aim in which we examine variability in children’s incoming and outgoing talk with teachers and peers, we ran a series of six unconditional multilevel models at each of the 8-sensing observation timepoints. Dependent variables for these models and Intraclass Correlation Coefficient (ICC) results are summarized in Table 1. To examine associations between child- and classroom-level factors and children’s talk, we fitted a series of conditional trajectory models accounting for the nesting of children within classrooms with incoming and outgoing talk and conversational turns between target children and their teachers and peers as the dependent variables. Results show within and between classroom variability in children’s teacher and peer talk. With respect to Aim 2, a variety of factors are associated with the talk children experience including their caregiver’s education and characteristics of their classroom social network (see Table 2). This work highlights individual differences in preschool children’s talk experiences, and the value of sensing observation systems to objectively measure multiple children’s talk in the classroom simultaneously over time.

Paper #2
Capturing Attention in Real-Time: Variability across Instructional Contexts and Individual Characteristics
Author information	Role
Zoe Mao, UCLA, United States	Presenting author
Fang Yu Chang, UCLA, United States	Non-presenting author
Agatha Lenartowicz, UCLA, United States	Non-presenting author
Jennie K. Grammer, UCLA, United States	Non-presenting author
Abstract
There is great variability in children’s learning experiences as a function of individual characteristics and the classrooms they attend (Myers & Pianta, 2008). These experiences shape children’s development of cognitive skills, such as attention and executive functions (Grammer & Torres, 2021). However, it is challenging to capture the bidirectional experiences and examine their influences on cognitive processes, which may not be directly observable. To examine interactions between students and variations in instruction, we conducted an experimental study and assessed student attention in a semi-naturalistic setting, using observational rating and electrophysiological (EEG) measures. Existing protocols for observing student attention are often used in siloed disciplines (clinical, educational, developmental) and based on discontinuous measurement of behaviors (e.g., partial interval). We developed a new observational protocol – Student Attention Tracking (SAT) – to capture the fluctuations in attention-related behaviors by coding continuously throughout a learning activity and by integrating multidisciplinary perspectives of active/passive engagement, fidgeting, off-task verbalization, and gaze (Farley et al., 2013; Rapport et al., 2009; Shapiro, 2004). In this presentation, we address two complementary aims: 1) to assess SAT reliability by comparing the observational ratings against other measures of attention and EF, and 2) to examine differences in children’s SAT ratings across different instructional contexts and child characteristics. Children aged 6-11 (N=80; mean age=8.35; 37.5% female; 32.5% ADHD) participated in four one-on-one science learning activities that varied in who led instruction and mode of delivery – in-person instruction, online instruction, video watching, and independent activity. SAT observational ratings assessed participants’ attention, fidgeting, and off-task behaviors during the activities. Additionally, direct assessments included one spatial working memory and two executive functioning tasks, and standardized parent reports of children’s behavior included Conners3, BRIEF, and CBCL. Aim 1 Results Given the interconnected nature of the constructs measured (Barkley, 1997), strong associations between the measures will suggest good reliability of the new SAT protocol. After controlling for children’s sex at birth, age, and family SES, in multiple regression analyses, we found consistently strong associations between observational ratings and parent reports (attention: standardized β=-.28–-.34, ps<=.01; fidgeting: β=.23–.29, ps<.05; off-task: β=.22–.35, ps<=.05), both reflecting attention-related behaviors. Aim 2 Results To understand the influences of instructional activity (within-subjects) and child ADHD diagnostic (between-subjects) on observed attention, fidgeting, and off-task behaviors, we ran separate mixed-design ANOVAs (Table 1). We found significant main effects and interaction effects, indicating that children’s attention-related behaviors varied across activities and by ADHD diagnosis (Figure 1). Results indicate that instructional activity had a much larger effect on attention and fidgeting (ηp2=0.697, 0.647, respectively) than ADHD diagnosis (ηp2=0.196, 0.140, respectively). So different activities accounted for more than three-fifths of the variability in children’s attention-related behaviors. These results highlight the importance of factors within the control of educators – namely the type of instructions they choose – for shaping children’s attention and attention-related behavior. We will discuss these findings, and the implications of this work for teacher preparation and professional learning.

Paper #3
Child and Classroom Predictors of Children’s Sorting Patterns: Teachers’ Instructional Language and Children’s Cognitive Factors
Author information	Role
Amber Elizabeth Westover, UNC Greensboro, United States	Presenting author
Kaitlyn L. Tran, UNC Greensboro, United States	Non-presenting author
Peter A. Ornstein, UNC Chapel Hill, United States	Non-presenting author
Jennifer L. Coffman, UNC Greensboro, United States	Non-presenting author
Abstract
A rich literature documents age-related changes in children’s strategic memory behaviors across elementary school (Roebers, 2014). However, little is known about predictors of intra-individual change. Researchers have suggested the importance of cognitive factors such as metamemory and attentional processes for the development of these skills (Pressley et al., 1985). Recently, classroom experiences, specifically teachers’ metacognitively-rich instructional language, have been associated with children’s use of memory strategies (Ornstein & Coffman, 2020). Yet, few investigations have considered the role of child and classroom factors simultaneously. Moreover, no known studies have explored how these factors may predict patterns of growth in children’s memory skills over time. A sample of 75 children was followed across kindergarten and first grade. Children were asked to remember 16 pictures from 4 categories and their sorting (meaning-based grouping) was assessed (Adjusted Ratio of Clustering; Roenker et al., 1971). At both fall assessments, spontaneous sorting was measured during a baseline trial, followed by brief experimented-led training in sorting. Finally, a generalization trial was administered with new stimuli to assess uptake of training. In the winter and spring, only generalization trials were used. During kindergarten, children’s metamemory knowledge was assessed using memory scenarios and parents rated their children’s attentional focusing and inhibitory control. Children’s classrooms were observed to evaluate teachers’ use of metacognitively-rich instructional language (Cognitive Processing Language; CPL) during whole-class lessons. Teachers were identified as using high or low levels of CPL. A cluster analysis (Ward, 1963) was conducted to examine patterns of growth in children’s strategic sorting: five clusters were identified (Figure 1). There were marked differences in when each group adopted the sorting strategy. One group took it up during kindergarten (K Training Responsive, n=19), whereas others adopted it either the start of first grade (Post K Training Adoption, n=19), in response to first-grade training (First-Grade Training Responsive, n=11), or at the final timepoint (Post First-Grade Training Adoption, n=12). Notably, one group engaged in little-to-no sorting across the study (Low Strategic, n=19). A multinomial logistic regression was used to explore both classroom- and child-level predictors of group membership (Table 1). The two groups who adopted the sorting strategy early showed higher attentional focusing abilities and exposure to higher CPL in kindergarten, relative to their Low Strategic peers. Notably, the children who took up the strategy first (K Training Responsive) also demonstrated greater metamemory knowledge. Interestingly, the Post First-Grade Training Adoption group was more likely to have higher metamemory but had lower inhibitory control than the Low Strategic group. These findings underscore the combined role of classroom and child-level factors in differential patterns of children's memory strategy development. Children who responded to strategic training and adopted an organizational memory strategy early on demonstrated a combination of richer metacognitive classroom experiences and higher levels of individual abilities (attentional focusing and metamemory knowledge). Notably, child-level factors alone (e.g., metamemory) may be insufficient to support the development of these skills without supportive classroom environments. Thus, both classroom and child-level differences should be considered when exploring the development of memory skills.

Paper #4
“Inclusion” as a Sense of Community: Autistic and Nonautistic Students’ Interactions with Teachers and Peers
Author information	Role
Nicole Sparapani, UC Davis, United States	Presenting author
Cameron Alexander, UC Riverside, United States	Non-presenting author
Johanna Vega Garcia, UC Davis, United States	Non-presenting author
Sandy Birkeneder, UC Davis, United States	Non-presenting author
Abstract
Researchers, stakeholders, and policy makers emphasize the importance of inclusion within general education classrooms, yet little is known about successful inclusion for autistic learners (Biklen, 2020; Haegele & Maher, 2023). Recent studies have suggested that features of teacher-student interactions provide insight into successful inclusion, such as reciprocity between teachers and students, opportunities to collaborate with peers, and the individual variability in patterns that arise (Despois & André, 2024; Heller & Kern, 2021). Repeated positive or negative interactions between teachers and their students, for example, send messages to students that they do or do not belong in the classroom community (Ball, 2018). These public interactions also send important signals to peers about acceptance or rejection of others, with research demonstrating that peers are likely to hold inherent biases about neurodivergent students if their teachers do (Aube et al., 2021). This study utilized classroom video observations to identify patterns in interactions that promote or hinder inclusive communities. Participating students (n = 82; 41 autistic) and teachers (n = 22) were recruited for a 4-year project (R324A210288) examining teacher-student interactions within TK–5th grade general education mathematics and literacy lessons. The research team administered a battery of measures, teachers completed questionnaires, and video observations were collected throughout the year. Trained observers coded teacher-student interactions using the Class-MAP system (Sparapani et al., unpublished) with high interrater agreement (percent agreement >80%). We report preliminary findings on a neurodiverse subsample of 63 students (Mage = 6.56; SD = 1.56; 32 autistic) and their 18 teachers within a 30-minute sample of mathematics instruction. The sample was racially, ethnically, and linguistically diverse. Grade level, age, sex, and cognitive skills were similar between groups. Teachers reported significantly higher rates of conflict (M = 2.27; SD = 1.11) with their autistic students, F (1) = 7.52, p < .01. We observed significant differences in the patterns of communication initiations between autistic and nonautistic students, with autistic students initiating less communication to show, F (1) = 4.60, p < .05, and give, F(1) = 8.15, p < .01, and engaging in fewer peer-to-peer interactions, F(1) 5.86, p < .05, than their nonautistic peers. Teachers responded to their autistic students more often (47%) than their nonautistic students (30%), F(1) 4.37, p < .05. We observed group differences in executive functioning skills (BRIEF), adaptive functioning (Vineland-3), joint attention (ESCS), and sensory processing skills (Sensory Profile-2). We also observed moderate, negative correlations between the frequency of students’ communication initiations with their sensory processing (r = –0.59) and executive functioning skills (r = –0.47) following FDR adjustments and controlling for cognitive functioning. These preliminary findings provide insight into the experiences of autistic learners within their classroom communities, while highlighting areas of need linked to communication initiations. Examining nuances in teacher and peer responses and initiations is our next step—information that will provide insight into the classroom community as an indicator of successful inclusion.

⇦ Back to session