Sequential analyses in coercive mother-child interaction: the predictability hypothesis in abusive versus nonabusive dyads

Abstract

Objective: A two-fold purpose guided the present study: 1) To test the sequential relationship between the child’s aversive behavior and both the predictability and the compliance episodes, as well as the sequential relationship between these two mothering episodes proposed by the new predictability hypothesis (Wahler, Williams, & Cerezo, 1990); 2) to explore whether or not these patterns are specific to these dysfunctional dyads by using a nonabusive comparison group.

Method: Fifty mother-child dyads, 25 abusive and 25 nonabusive, participated in this study. Lag sequential analyses were carried out on 302 hours of direct observation, 178 hours in the abusive group, and 124 in the nonabusive group.

Results: Results of the sequential analyses showed that the predictions derived from the compliance and predictability hypotheses were supported in both groups. The only difference between the groups lies in the new predictability hypothesis, so the two mothering reactions are not sequentially related in the nonabusive group.

Conclusions: This study revealed interesting findings which suggest that these patterns are also found in nonabusive mother-child interactions, but there is a qualitative difference between the groups.

Résumé

Objectif: L’etude avait deux buts : (1) vérifier la relation séequentielle entre le comportement déplaisant d’un enfant et les épisodes prévoyables et conformes que propose la nouvelle hypothese sur la capacité de prédire (Wahler, Williams, et Cerezo, 1990); ainsi que la relation entre deux épisodes de soins maternels; et (2) voir si ces liens sont spécifiques à ces deux dyades dysfonctionnelles, ceci en se servant d’un groupe de comparaison composé de personnes non abusives.

Méthode: L’étude comprit 50 dyades de mères et d’enfants, dont 25 avaient des relations abusives et 25 n’en avaient pas. On a observé les sujets pendant 302 heures (178 heures pour le groupe abusif et 124 heures pour le groupe non abusif).

Résultats: Les résultats démontrent que pour les deux groupes, on a pu prédire, tel le veut les hypothèses de conformité et de capacité de prédire. La seule différrence est au niveau de la nouvelle hypothèse sur la capacité de prédire, pour laquelle, dans le groupe non abusif, les deux réactions maternelles ne démontrent aucune association séquentielle.

Conclusions: Cette étude révèle des découvertes intéressantes suggérant que ces ensembles structurels sont aussi présents dans les interactions mère-enfant non abusives, cependant, il existe une différence qualitative entre les deux groupes.

Resumen

Objetivo: El presente estudio se llevó a cabo con dos objetivos: (1) Evaluar la relación secuencial entre la conducta aversiva del niño/a y episodios de predictibilidad y obediencia materna, ası́ como la relación secuencial entre estos dos episodios maternos, propuestos por la nueva hipótesis de la predictibilidad (Wahler, Williams, & Cerezo, 1990), y (2) explorar si estas pautas son especı́ficas de estas dı́adas disfuncionales, utilizando un grupo comparación de familias no-abusivas.

Método: Cincuenta dı́adas madre-hijo, 25 abusivas y 25 no-abusivas, participaron en este estudio. Se llevaron a cabo análisis secuenciales de retardos (lag) sobre 302 horas de observación directa, 178 horas en el grupo abusivo y 124 en el grupo no-abusivo.

Resultados: Los resultados de los análisis secuenciales apoyaron en los dos grupos las predicciones derivadas de las hipótesis de la obediencia y la predictibilidad materna. La única diferencia entre los grupos se produjo en la nueva hipótesis de la predictibilidad, pues en el grupo no-abusivo no se encontró una relación secuencial entre los dos episodios de interacción madre-hijo.

Conclusión: Este estudio tiene hallazgos interesantes que sugieren que estas pautas también se encuentran en las interacciones no-abusivas madre-hijo, pero que hay una diferencia cualitativa entre los grupos.

Introduction

IN THE AREA of conduct disordered children, research has shown a consistent relationship between irritable parent discipline practices, marked by parent-child coercive exchanges, and the development of conduct disorders in children. In the last decade, Wahler and Dumas (1986) developed two specific hypotheses which have guided further empirical studies: the compliance hypothesis and the predictability hypothesis. According to the compliance hypothesis, which represented the conceptualization for instructional episodes of Patterson’s coercion theory (Patterson, 1982), the mother’s instructional behavior generates a setting in which the child’s oppositional and aversive behavior is successful in terminating the maternal demand; thus, the mother escapes from the aversive situation by her “giving in” to the child’s opposition. The predictability hypothesis proposes that inattentive and erratic maternal behavior sets an interactional context for the child that is characterized by unpredictability. The escalating phenomenon which occurs in coercive exchanges is rewarding because the child’s deviant behavior is functional in obtaining a more predictable mother response. These two hypotheses found correlational support Dumas and Wahler 1985, Wahler and Dumas 1986.

The important issue related to the directionality of these events in the stream of interactional behavior was supported by strategies of sequential analyses in a further study (Wahler, Williams, & Cerezo, 1990). In the latter, an unplanned and interesting finding regarding the sequential relationship between the two mothering episodes led the authors to propose a “new predictability hypothesis” as “a process in which maternal compliance fosters a lock-step pattern of interpersonal aggression, initiated by a mother’s failure to enforce her demand” (p. 406). Unfortunately, the absence of a comparison group of nonclinical children in that study prevented the testing of whether or not these patterns were specific to clinical mother-child dyads.

In the area of child physical abuse and family interaction, these findings are extremely relevant in understanding the process of relationship dysfunction which occurs in maltreating households. Abused children live in highly conflictive families characterized by coercive exchanges and dysfunctional interaction (for a review, Cerezo, 1997). From this view, abusive parenting represents the very extreme of a continuum of dysfunctional socialization practices where mutual coercion entrapment is part of daily life Ammerman 1990, Cerezo 1992, Wolfe 1987. For instance, Reid, Taplin, and Lorber (1981) compared nonabusive, distressed-nonabusive, and abusive families and reported that, in abusive families, aversive maternal behavior, such as disapproval or reprimand, significantly increased the likelihood of accelerating the child’s aversive response instead of stopping this behavior. This kind of interactive sequences were used as an operational index of the escalation phenomenon. Further studies replicated these findings (i.e., Loeber, Felton, & Reid, 1984).

When child physical abuse is focused on as the outcome of the extreme lack of parental skills and dysfunctional parenting, the children have an elevated risk of psychopathology and in particular, of antisocial behavior and conduct disorders. In fact, the resemblance of abused children to conduct-problem children in their negative behavior at home has found empirical support Loeber et al 1984, Reid et al 1981, Whipple and Webster-Stratton 1991. These findings suggest the continuum of parental socialization practices as a connecting point between the research area of conduct disordered children and the research in child physical abuse. In this context, the unpredictable setting for the child’s development that is associated to an erratic parental discipline represents a central topic for both areas. Thus, the relevance of the inconsistent discipline as an important factor that characterizes abusing parental practices has been underscored by pioneering authors in the field (i.e., Parke and Collomer 1975, Young 1966). This interactional pattern in abused children has been empirically supported in observational studies (i.e., Cerezo and D’Ocon 1995, D’Ocon 1994).

According to the general previous findings and particularly the findings regarding the sequential relationships reported by Wahler and colleagues (1990) with conduct-disordered children, a study was designed which specifically focused on mother-child dyads with a history of physical and/or emotional child abuse. A two-fold purpose guided the present study. On the one hand, the goal was to confirm the predictions regarding the sequential relationship between the child’s aversive behavior and both the predictability and the compliance episodes, as well as the sequential relationship between these two mothering episodes proposed as the new predictability hypothesis. On the other hand, a second goal was to explore whether or not these patterns were specific to these dysfunctional dyads by using a nonabuse comparison group.

The hypotheses regarding the sequential relationship were formulated as follows. According to the compliance hypothesis, given a child’s deviant behavior, the conditional probability of maternal compliance episode was predicted to be significantly higher after that event than before. In terms of the predictability hypothesis, conditional probability of a maternal indiscriminate episode after a child’s aversive behavior was predicted to be significantly lower than before this event. Finally, an event sequential relationship between indiscriminate mothering episodes and the mother’s failure to enforce her demand on the child was expected to support the new predictability hypothesis. Regarding comparison between the two groups, abuse and nonabuse, the present study should be considered as exploratory because of the lack of previous findings involving these mothering episodes and their sequential relationships with child’s behavior in nonabuse dyads.