The Efficacy of Self-Directed Speech Training on Executive Functioning: A Multiple Single-Case Pilot Study

Developmental language disorder (DLD) is characterized by difficulties in developing age-appropriate expressive and receptive language skills that are not associated with known biomedical causes. In addition to language problems, many children with DLD have difficulties with executive functioning (EF). It has been suggested that these EF difficulties can be explained in terms of children’s language problems. Indeed, it has been argued that language, in the form of self-directed speech, becomes a tool for self-regulation by allowing individuals to verbalize and thereby regulate their thoughts and behavior. The training Talk to Think aims to improve EF in children with DLD by promoting self-directed speech. The effectiveness of this training has been assessed in one study using a wait-control design. However, that study was conducted exclusively at the group level and therefore could not take into account the high degree of heterogeneity within the DLD population. We therefore relied on a multiple-baseline single-case design. EF was measured in two distinct ways. First, puzzles from the Tower of London task were administered weekly during both the five baseline measurements and the five training-phase measurements. Everyday EF was assessed pre- and post-training using the Dutch version of the Behavior Rating Inventory of Executive Function. Children’s speech during the Tower of London was transcribed from video recordings and coded according to self-directed speech coding standards. For each participant, we computed the number of task-relevant self-directed utterances per Tower of London item for each session. The collection contains, for each session, both the trial-level and participant-level data for the Tower of London and the private-speech coding. In addition, it contains the fully completed BRIEF questionnaires both prior to and after training, as well as the item-level data for the Raven’s Coloured Progressive Matrices and the word meaning structure test, both administered prior to training. The R scripts used to compute descriptives, set up models, and generate figures are also included. The names in the collection are fictitious.