The effect of relational training on intelligence quotient: A case study

Full reference: Vizcaíno-Torres, R.M., Ruiz, F.J., Luciano, C., López-López, J.C., Barbero-Rubio, A., & Gil, E. (2015). The effect of relational training on intelligence quotient: A case study. Psicothema, 27(2), 120-127.

Study type: Single-case experimental design (N=1) based on Relational Frame Theory

Background and objectives

Relational Frame Theory (RFT) is a functional-contextual approach to complex human behavior that holds two main premises with important implications for the training of linguistic and cognitive abilities. The first premise is that fluency and flexibility in different patterns of arbitrarily applicable relational responding, or relational framing, underlie language and cognition. Relational framing means responding to one event in terms of another where the relationship between the two events is not based on nonarbitrary features but on arbitrary relational cues (e.g., "same as"). Fluency in relational framing refers to the ease of deriving relations, whereas flexibility refers to the ability to relationally frame stimuli under different relational cues, contexts, and formats.

The second premise is that all patterns of relational framing are generalized operant behaviors that are learned through multiple exemplar training (MET). MET consists of a process in which multiple examples of mutual relations, combinatorial relations, and transformations of functions with a particular pattern of relational framing are provided, using multiple sets of stimuli. Empirical evidence supports that relational frames such as coordination, opposition, or comparison can be trained through MET. Recent empirical evidence has found that abilities in relational framing correlate with performance on standardized intelligence tests and specific cognitive skills. Indeed, psychometric measures of intelligence can be deconstructed in terms of the specific relational responses necessary to solve their items.

The current study aimed at adding additional evidence to the improvement of intelligence measures by training fluency and flexibility in relational framing according to coordination, opposition, and comparison in a normally developing 4-year-old child. Unlike the only study conducted with normally developing children, the actual protocol involved the establishment of relational framing according to opposition and comparison because these repertoires were absent, and included fluency and flexibility training both in-session and in daily interactions.

Method

Participants / Subjects / Models

CR, a boy who was 3 years and 8 months old at the beginning of the study, was the participant of this case study. According to his parents' reports, he was a healthy, happy, and occasionally shy boy. He was at the appropriate stage of social development and grade level for his age although he did not stand out in any particular subject. He was enrolled in Preschool during the period of the study.

CR's mother, a first year Ph.D. student, administered all tests and relational trainings. The study lasted approximately five months, beginning when the child was 3 years, 7 months and 23 days old, and finished when he was 4 years and 11 days old. Ten sessions were conducted in total: five to administer the assessment instrument and five to apply the training protocol.

Experimental design

The design of this study was N=1. The main dependent variables were the scores obtained by CR in the McCarthy's Aptitudes and Psychomotricity Scale (MSCA). This instrument was administered before and after the introduction of the independent variable: a training protocol in relational framing. The independent variable consisted of six sequentially structured training phases. Phases 1, 3, and 5 were dedicated to the evaluation, training, and testing of relational framing according to coordination, opposition, and comparison, respectively. In Phases 2, 4, and 6, fluency and flexibility were promoted in the use of the aforementioned relational frames (Phase 2: coordination; Phase 4: opposition and coordination; Phase 6: comparison and opposition).

Materials and apparatus

A 50cm × 30cm × 10cm box was used as a physical barrier between the researcher and the child at the time of touching or naming the stimulus. On some occasions, stimuli were presented using a computer program (Microsoft PowerPoint®) and an Apple laptop (MacBook Pro, 15.4 inches). A video camera (Sony Handycam HDD LCD Wide®) was used to record sessions and the child's responses. The stimuli used in different phases varied according to the objective of each phase. In Phase 1, five unknown objects were used (padlock, thimble, screw, tape measure, corkscrew) with their corresponding Spanish names. In Phase 3, for training nonarbitrary opposite relations, concrete stimuli were used (potatoes, glasses of water, porcelain containers, boiled eggs, sponges, kitchen napkins, cotton balls) dimensioned according to opposite characteristics (hot-cold, dry-wet, hard-soft). For arbitrary opposite relations, drawings of colored glasses and boxes of different colors were used, as well as surprise gift boxes. In Phase 5, for training nonarbitrary comparative relations, wooden towers with different numbers of pieces, bottles, paint brushes, stickers, and containers were used. For arbitrary comparative relations, cards of different colors, balls, drawings of bags, trash cans, and opaque containers were used.

Procedure

The training protocol was based on guidelines presented in previous studies: (a) evaluation of the relational repertoire, (b) sequential and errorless training, (c) initially training with nonarbitrary relations and then with arbitrary relations, (d) use of several sets of stimuli with several dimensions and functions, (e) use of different training formats, and (f) assessment of transformation of functions through mutual and combinatorial entailment.

During all phases, CR's correct responding was followed by positive social feedback and intermittently by a happy face sticker. He was told that he could exchange the stickers for some prizes at the end of the day. Incorrect responding was followed by the experimenter saying "No, CR, that is not right" or "No, that is not correct," with a sad face sticker also added intermittently. The stickers and social feedback were faded across trial phases: in Phase 1, they were provided every three correct responses; in Phase 3, every six correct responses; and in Phase 5, every eight correct responses.

Phase 1. Evaluation of relational framing according to COORDINATION. This phase began when CR was 3 years, 8 months, and 4 days old. First, a generalized and delayed naming test (auditory-visual mutual relations) was conducted with receptive and productive trials using five unknown objects. In receptive trials, CR was informed of the name of an object and was asked to produce that word. Then, after 60 minutes, he was asked to take the object from among an array of several stimuli. In productive trials, CR was informed of the name of an object and was asked to produce the word after a 60-minute delay. CR responded correctly to all trials. Subsequently, four visual-visual conditional discriminations (B1-A1, C1-A1, B2-A2, C2-A2) were trained in a many-to-one matching to sample procedure with three comparisons. Afterward, mutual (A1-B1, A1-C1, A2-B2, and A2-C2) and combinatorial relations (B1-C1, C1-B1, B2-C2, and C2-B2) were evaluated. The procedure used 80 trials in total (64 training trials and 16 test trials, two per relation) in a single session. CR responded correctly to all test trials.

Phase 2. Promoting fluency and flexibility in relational framing according to COORDINATION. Diverse opportunities were generated in everyday language activities with new examples of arbitrary relations, and responses were requested in different contexts and changing functions. Eight stories were used with a total of 8 mutual and 12 combinatorial relations. CR responded correctly to all 20 test trials.

Phase 3. Evaluation and training of relational framing according to OPPOSITION. This phase began when CR was 3 years, 11 months, and 5 days old, and was conducted intensively for a period of 150 minutes, with several 10-15 minute breaks. First, an evaluation including five mutual entailment trials with arbitrary relations was conducted to explore CR's abilities in relational framing through opposition. CR responded correctly to only one trial.

Given that this repertoire was absent, a MET with nonarbitrary opposite relations was conducted. This training consisted of 60 trials with three dimensions and seven sets of stimuli. Transformation of functions through mutual entailment was trained first. The first trial of a dimension was presented twice. This stage ended when CR responded correctly to 8 consecutive trials.

Subsequently, CR was exposed to similar training with three stimuli for training mutual and combinatorial entailment. Finally, when CR responded correctly, the three stimuli were presented together and the corresponding transformation of functions was evaluated. When CR made a mistake, he was allowed to touch the stimulus to obtain feedback. This stage finished when CR responded correctly to 8 consecutive trials. Overall, CR responded correctly to 37 of the 47 training trials.

Afterward, a MET of mutual and combinatorial relations was conducted with arbitrary relations (13 trials in total). This training was similar to the previous one, but trials were presented on the computer screen and involved two dimensions (full/empty, sweet/salty) and three sets of stimuli. Finally, CR was exposed to a test of arbitrary mutual and combinatorial relations consisting of 12 trials (8 mutual and 4 combinatorial) with two sets of stimuli. CR passed the test by responding correctly to 11 of the 12 trials.

Phase 4. Promoting fluency and flexibility in relational framing according to OPPOSITION and COORDINATION. Similar to Phase 2, fourteen stories that led to 36 trials involving both mutual and combinatorial relations were used in daily interactions. CR responded correctly to 95% of the trials with opposite relations and to 100% of the trials with coordination.

Phase 5. Evaluation and training of relational framing according to COMPARISON. This phase began when CR was 3 years, 11 months and 20 days old, and was conducted in a single session that lasted 3 hours and included 114 trials and several 10-15-minute breaks.

First, an evaluation including eight trials with arbitrary relations was conducted to explore CR's abilities in relational framing through comparison relational cues (e.g., "more than," "less than"). Three sets of stimuli were used. CR responded incorrectly to all trials.

Once it was shown that this repertoire was absent, a MET with nonarbitrary mutual and combinatorial relations was conducted (85 trials) with other sets of stimuli. Two dimensions (volume and size) and seven sets of stimuli were used. A typical trial proceeded as follows: Two towers (A>B) with a different number of pieces of wood were presented and CR was asked: "Which of them has more pieces? Which has fewer?" Progressively, the differences between the towers were minimized throughout 3 trials while the same questions were asked. Later, a third tower (A>B>C) was presented and CR was asked for the combinatorial relations (A>C and C<A). If CR responded incorrectly, he was allowed to count the pieces of the tower. When CR produced 15 correct consecutive responses, a MET with arbitrary relations (21 trials) commenced with four dimensions and six series of stimuli. A typical trial proceeded as follows: Cards identical in size but different in color were used. CR was informed about arbitrary quantitative relations between the cards (e.g., "With the red card you can buy less ice cream than with the black card"). The experiment required CR to respond to questions about these relations and then about derived relations (combinatorial relations). If the child responded incorrectly, the relation "If A is... and A is more/less than B, then B is..." was trained explicitly.

After this intensive training, CR responded correctly to 8 of the 10 trials of the test with arbitrary relations involving the same sets used in the previous evaluation of this repertoire.

Phase 6. Promoting fluency and flexibility in relational framing according to COMPARISON and OPPOSITION. Twenty-four new stories involving 56 comparison and 16 opposition trials were presented on a day-to-day interactions basis. He responded correctly to 92.9% of comparison and 93.8% of opposition trials.

Data analysis

CR's direct scores on the MSCA at pre-intervention were interpreted according to the scales for 3 years and 4 months to 3 years and 9 months when he was 3 years and 7 months old. Direct scores at post-intervention were interpreted according to the scales for 3 years and 10 months to 4 years and 3 months when he was 4 years old. Ninety percent confidence intervals were obtained from the MSCA manual.

The integrity of the protocol administration (Phases 1, 3, and 5) was measured by means of inter-observer agreement (agreement among two independent observers divided by the sum of agreement and disagreement, multiplied by 100). Both observers claimed 100% agreement in evaluation, training, and tests regarding the correct presentation of the trials, identification of CR's responses, and adequate feedback provision.

Results

At pre-intervention, CR obtained average levels adjusted to the norm in all areas (T scores between 40 and 59) and in the General Cognitive Index (GCI = 106, 90% CI [99, 113], percentile 65, and a direct score of 80). At post-intervention, CR showed increases in the scores on all areas, especially in the verbal area where he obtained a T score of 49 at pre-intervention (90% CI [44, 54]) and of 66 at post-intervention (90% CI [61, 71]). CR obtained a very high level in the GCI, scoring 131 points (90% CI [124, 138], percentile 97, and a direct score of 140).

The training was effective in establishing relational responding in OPPOSITION and COMPARISON frames as well as in promoting fluency and flexibility in all three types of trained relations. Performance data during training showed that in Phase 1 (coordination), CR responded correctly to 5 of 5 trials, followed by 16 of 16 test trials; in Phase 2 (fluency in coordination), he responded correctly to 8 of 8 trials; in Phase 3 (evaluation and training of opposition), CR initially responded correctly to 1 of 5 evaluation trials, after 73 MET trials he responded to 11 of 12 trials on mutual relations tests and 12 of 12 on combinatorial relations tests; in Phase 4 (fluency in opposition and coordination), he responded correctly to 95% of opposition trials and 100% of coordination trials; in Phase 5 (evaluation and training of comparison), CR initially responded correctly to 0 of 8 trials, after 106 MET trials he responded to 8 of 10 test trials; in Phase 6 (fluency in comparison and opposition), he responded correctly to 92.9% of comparison trials and 93.8% of opposition trials.

The most significant changes were observed in the verbal area and in the GCI. The 25-point increase in the GCI (from 106 to 131) represents a change above 1.5 standard deviations. Anecdotal reports from CR's teacher and family were consistent with the improvement found in the MSCA (e.g., CR performed better at school, showed better verbal understanding than his schoolmates).

Discussion and conclusions

CR entered the study performing at average levels in the MSCA. After the implementation of the training protocol, he showed improvements in all areas of the MSCA. These changes were especially relevant in the verbal area and in the GCI, where CR showed a 25-point increase (from 106 to 131). Consistent with previous studies with normally developing children, METs were effective in establishing relational framing in accordance with opposition and comparison. However, in this study both relational frames were trained with the same participant, and the number of sessions necessary to establish them was reduced. As suggested in the literature, this might be due to the incorporation of diverse stimuli with different functions in addition to the initial training in nonarbitrary relations, as conducted in previous studies, which quickly facilitated the abstraction of the relational cues. A novelty of the current study was the incorporation of daily natural interactions to promote fluency and flexibility in relational framing.

However, the current study has all the limitations of N=1 designs. In addition, some other limitations are noteworthy. First, the experimenter was CR's mother. This made the protocol implementation less intrusive and facilitated the day-to-day interactions to promote fluency and flexibility in relational framing, but her behavior could be more strongly affected by the expectations of the experimenter. For instance, it could be that the mother inadvertently provided some cues that would have facilitated CR's responses both during the training and tests. Second, the training protocol used was formed of coordination, opposition, and comparison framing. Further studies should study the effect of a protocol training including relational framing according to relations such as distinction, hierarchy, spatial, causality, and deictic. Third, this study lacked a control participant that would allow controlling for the child interactions that occurred throughout the five months of training.

In conclusion, the current case study provides promising evidence of the effect of an RFT-based training in improving the linguistic and cognitive abilities categorized within the construct of intelligence. In this sense, this type of studies might change the way most psychologists conceive intelligence as a relatively stable construct across lifetime. However, better controlled studies are required to analyze the potential of these RFT-based trainings.

Significance and Contribution

This case study provides empirical evidence for the potential of behavioral training based on relational framing principles to improve measured cognitive abilities and intelligence. The research is significant because it challenges traditional notions of intelligence as a stable construct, suggesting instead that fundamental cognitive abilities can be substantially improved through specific behavioral trainings. The findings contribute to understanding the underlying mechanisms of relational behavior and its role in cognition, with practical applications in educational and clinical contexts, particularly for children with developmental delays or learning difficulties. The 25-point improvement in the General Cognitive Index observed in this normally developing child demonstrates that cognitive plasticity can be systematically fostered through well-designed behavioral training, contributing to a broader understanding of how intervention science can enhance cognitive and linguistic repertoires in developing children.