Computerized Teaching of Bidirectional Naming

Abstract

Teaching Bidirectional Naming (BiN) to typically and atypically developing children has been extensively investigated using procedures that rotate multiple exemplars of multiple stimuli from the same class and multiple instructions. Interventions are usually applied by an experimenter who manually presents all the stimuli, instructions, and consequences, in addition to managing randomization and registering responses. Manual application of interventions is not only prone to error but also does not allow experimental control of relevant variables such as session duration, response times, and post-reinforcement pauses. However, to date, few studies have reported employing computerized procedures for individualized instruction, and only two published studies were specifically designed to analyze its implications: a brief report by Lee et al. (2021) and an original research article written by Lee et al. (2025). The present study described and evaluated the effects of a computerized version of the Multiple Exemplar Instruction (MEI) protocol on BiN emergence in six Brazilian children. Participants were five typically developing children and one child diagnosed with an intellectual disability. The intervention followed established procedures in the literature, rotating four types of trials: (1) matching-to-sample (MTS) with a compound model (an identity MTS combined with the spoken name of the object), (2) auditory-visual MTS, (3) tact with intraverbal, and (4) pure tact. Participants were evaluated in BiN probes before and after MEI sessions on the tasks: a) intraverbal tacts, and b) visual-auditory MTS, which were also computerized. The BiN assessment was performed using both the original set and sets with novel stimuli, in order to verify performance generalization. Tests were always conducted after exposure to the stimulus names during compound MTS teaching blocks, constituting a sort of Naming Experience. All participants improved their performance in teaching and test sessions, and two developed BiN after the protocol application, achieving higher than 80% of correct responses in the two emergent behaviors tested. The results demonstrate the extension of the viability of MEI computerization for BiN teaching to neurotypical Brazilian learners. Other potential benefits of this form of application were: 1) shorter session duration than during manual application, 2) more precise experimental control for variables such as order randomization for responses, stimuli and exemplars, response latency and accuracy recording, 3) easier and potentially cheaper experimenter training, and 4) potential for large-scale use in applied scenarios, such as school settings. Limitations included the absence of control for the passage of time and the reliance on a single data point of BiN evaluations. Future studies should employ multiple baselines across participants design and have a stability criterion for test outcomes to improve experimental control and generalization of results.