Environmental transmission of generalized anxiety disorder from parents to children: worries, experiential avoidance, and intolerance of uncertainty.

Generalized anxiety disorder (GAD) runs in families. Building on recent theoretical approaches, this review focuses on potential environmental pathways for parent-to-child transmission of GAD. First, we address child acquisition of a generalized pattern of fearful/anxious and avoidant responding to potential threat from parents via verbal information and via modeling. Next, we address how parenting behaviors may contribute to maintenance of fearful/anxious and avoidant reactions in children. Finally, we consider intergenerational transmission of worries as a way of coping with experiential avoidance of strong negative emotions and with intolerance of uncertainty. We conclude that parents with GAD may bias their children's processing of potential threats in the environment by conveying the message that the world is not safe, that uncertainty is intolerable, that strong emotions should be avoided, and that worry helps to cope with uncertainty, thereby transmitting cognitive styles that characterize GAD. Our review highlights the need for research on specific pathways for parent-to-child transmission of GAD.

Environmental transmission of generalized anxiety disorder from parents to children: worries, experiential avoidance, and intolerance of uncertainty

Defined by excessive levels of anxiety and worry,[1] generalized anxiety disorder (GAD) is a common and debilitating disorder among adolescents and adults.[2-6] Lifetime prevalence ranges from 2% to 6%,[4-6] and onset occurs between the late teens and twenties.[7,8] Early-onset forms run a chronic course and are resistant to treatment.[8,9]

Like other anxiety disorders, GAD runs in families.[10,11] Its presence in first-degree relatives predicts a twofold increase in the prevalence of anxiety/internalizing disorder and a five- to sixfold increase in the prevalence of GAD in other family members, suggesting some specificity in familial loading.[11] Genetic heritage from GAD parents, namely, a general disposition toward affective psychopathology, accounts for a significant but moderate (0.30 to 0.38) part of the variance in the overlap in GAD between parents and children.[10,12,13] Environmental influences should explain the remaining variance and the observed specificity of GAD transmission.[11]

This review focuses on the environmental transmission of GAD from parents to children. In contrast to adulthood years, where a stronger effect for nonshared than shared environment[14,15] was observed, in childhood years, a more pronounced contribution of shared environment to GAD was shown.[16,17] However, like other anxiety disorders, neither genetic dispositions nor shared/nonshared environmental components solely determine GAD in children.[16,17] As stressed by theoretical accounts of anxiety, it is the multiple interactions between risk and resilience factors at the genetic and environmental levels that determine the final adaptation in children.[18-21] Thus, environmental influences related to having a GAD parent are one of the many risk factors that probabilistically determine childhood GAD together with other risk or resilience-promoting aspects of the child and environment.

A better understanding of parent-to-child transmission of GAD is essential for devising effective strategies to prevent anxiety in children of GAD parents and for the treatment of children who meet GAD criteria. Considering the high prevalence,[2-6] the chronic and debilitating course of GAD,[8,9] and its economic burden to society,[22] efforts to reduce intergenerational transmission in families where one or both parents have GAD are highly valuable.

What distinguishes GAD from other anxiety disorders is not only the “generalized” nature of anxiety (eg, social, separation, somatic, performance, environmental threats, future), but also the specific way it is expressed via excessive worry due to anticipation of possible future threats.[23,24] The current review addresses both of these features. We first focus on the transmission of high levels of nonspecific trait anxiety from GAD parents via modeling, verbal transfer of anxiety, and parenting. More specifically, we address verbal and nonverbal pathways for a child's learning of fear/anxiety from parents in addition to parenting behaviors that may serve to maintain a child's fear/anxiety and avoidance. These three potential pathways for parent-to-child transmission of anxiety have been tested over a range of anxiety domains, but not yet investigated in GAD transmission. We therefore refer to these pathways as nonspecific (to GAD). We then address the second distinguishing feature of GAD, namely excessive worry and its intergenerational transmission. In regard to specific GAD transmission, we discuss children's learning of worrying from parents as a way of coping with high intolerance of uncertainty and experiential avoidance.

Nonspecific pathways in the parent-to-child transmission of anxiety

Verbal and nonverbal pathways for parent-to-child transmission of fear/anxiety and avoidance

Theoretical accounts on environmental transmission of fear/anxiety from parents to children have been largely conceptualized within the framework of fear acquisition models and social-learning theory.[25-27] Fears can be acquired indirectly from others via nonverbal and verbal transmission pathways. As GAD parents are, by definition, more likely to experience and express excessive anxiety to potentially threatening stimuli in daily life, verbal and nonverbal pathways for fear acquisition provide a useful framework to understand the environmental transmission of GAD from parents to children in daily interactions.

Nonverbal pathways for parent-to-child transmission of fear/anxiety and avoidance

Nonverbal fear acquisition involves learning from observation of others' reactions to ambiguous stimuli[28] (also referred to as vicarious/observational learning or modeling) and has been demonstrated as early as the end of the first year of life.[29-34] This early form of modeling, also known as social referencing (SR), refers to infants' use of others' reactions in response to novel stimuli to determine their own reactions to those stimuli.

SR studies explored the effect of parental fear/ anxiety expressions on infants' reactions to ambiguous stimuli (most often strangers, ambiguous toys, or a visual cliff) to shed light on early nonverbal transmission of fear/anxiety and avoidance in typical development, de Rosnay and colleagues compared the effect of maternal anxious (trained) versus nonanxious signals to strangers in a SR paradigm.[30] In this paradigm, a stranger engages the parent in an interaction while the child observes the interaction. Next, the stranger gradually approaches the infant and picks him/her up. Temperamentally inhibited infants who witnessed their mothers react in an anxious (vs nonanxious) way to strangers were found to be more fearful and avoidant during their subsequent interaction with strangers. In their famous visual cliff experiment, Sorce and colleagues demonstrated that none of the infants dared to cross to the deep end of a visual cliff when parents expressed fear in this situation, whereas most children crossed following their parents' happy facial expressions.[35] Rosen and colleagues showed that infants are less eager to interact with novel toys when their parents react with fear (vs joy).[36] Taken together, evidence supports a causal effect of parental anxious/fearful signals on child acquisition of fear/anxiety and avoidance of ambiguity, across the domains of social and situational anxiety in typical development at the end of the first year.

From toddlerhood onwards, relatively less is known on observational learning of fear and avoidance of ambiguity from parents. Gerull and Rapee compared toddlers' reactions to fear-relevant objects (a rubber snake and spider) when mothers expressed negative (fear-disgust) versus positive (happy-encouraging) emotions.[37] Toddlers showed more fear and avoidance of toys paired with the maternal negative emotions. Dubi and colleagues replicated this finding, also showing that child temperament does not moderate the influence of maternal verbal threat expressions on a child's behavior.[38] Dunne and Askew[39] investigated modeling of fear of unknown animals in an experimental paradigm where 6- to 10-year-olds were presented with pictures of maternal happy (vs fearful) facial expressions paired with novel animals. Children reported stronger fear of novel animals that were paired with the mother's fearful (vs happy) face. In a study investigating observational learning of panicrelevant escape and avoidance from parents, Bunaciu and colleagues used an experimental paradigm where 12-year-old female adolescents observed their parent performing a hyperventilation exercise.[40] In one condition, the parent completed the exercise; in the other, he/she modeled escape by discontinuing it. The findings revealed that participants who watched their parents quit the exercise themselves quit the hyperventilation exercise earlier. To summarize, the available evidence from community samples in infancy and childhood years reveal a causal influence of parental fearful and avoidant signals in the acquisition of fear, anxiety, and avoidance across the domains of situational and physical anxiety.

Murray and colleagues proposed that the emergence of SR and stranger anxiety at the end of infancy may mark a sensitive period for the parent-to-child transmission of social anxiety.[31] In a longitudinal comparison between infants of parents with versus without social anxiety disorder (SAD) in the stranger SR paradigm, they demonstrated that temperamentally inhibited children of socially anxious mothers become increasingly avoidant of strangers from 10 to 14 months, possibly as a result of exposure to parental expressions of anxiety. In a later replication and extension of this paradigm, Aktar and colleagues tested this direct link between parental expressions of anxiety and children's fear/ avoidance in a longitudinal design.[29] Similar to earlier evidence,[30,31] direct links were found between more expressed anxiety in parents and more avoidance of novel stimuli in temperamentally inhibited children across social (a stranger) and nonsocial (a robot) SR contexts.[29] The findings revealed that how much parents express anxiety matters, rather than their lifetime anxiety diagnoses. To summarize, the findings consistently reveal that observational learning from parental expressions of anxiety in infancy may lead to increased avoidance of ambiguity, especially in temperamentally fearful children of parents with anxiety disorders.

The findings from Aktar and colleagues[30,31] constitute the only longitudinal evidence available that extended the study of observational learning beyond infancy to study the parent-to-child transmission of anxiety in parents with (vs without) anxiety disorders.[41,42] In contrast to experimental studies (summarized above) revealing a causal influence of parental anxiety signals when children encounter an unknown stimulus (like a novel animal), findings after infancy from this sample reveal that parental expressions of anxiety no longer directly predict a child's reactions in response to strangers and mechanical robots. Nevertheless, the longitudinal investigation of SR processes in this sample[41,42] revealed significant longitudinal links between more expressed anxiety in parents and more fear/avoidance in children in the period between infancy and early childhood, but only in the case of additional vulnerabilities resulting from an inhibited temperament or from severe anxiety disorders in parents.[41,42] Taken together, findings suggest that observational learning from parents' earlier expressions of anxiety in SR situations may have immediate effects on a child's anxious/avoidant behaviors in infancy, and prolonged effects on later child behavior.

Verbal pathways for parent-to-child transmission of fear/ anxiety and avoidance

Verbal fear acquisition refers to learning of fear/anxiety and avoidance via parental verbal transfer of threat information (also called instruction learning). In his seminal work on fear acquisition, Rachman[25] highlighted verbal transmission of fear/anxiety as one of the main pathways through which parents transmit information about potential dangers to offspring. In contrast to the hypothesized importance of verbal pathways for parent-to-child transmission of anxiety and to available evidence on a causal role of the verbal information pathway in children's acquisition of fear from others,[43] only a few studies have investigated the verbal transmission of fear/anxiety.

Experimental studies investigating the causal role of verbal threat information on child fear acquisition have predominantly focused on childhood years. Muds and colleagues studied the effect of parental verbal threat information on 8- to 13-year-old children's fear of unknown animals.[44] Parents were provided with positive, threat, or neutral information about an unknown animal and instructed to communicate this information to their children. They found that the narratives of parents who were provided with threat (versus positive) information were indeed more negative and threatening, giving rise to children believing the animals were dangerous. In addition to self-reported fear beliefs, Bosnians and colleagues[45] and Remmerswaal and colleagues[46] measured children's behavioral avoidance in this experimental paradigm using a touch box said to contain the animals their mothers provided threat information about. Threat information caused behavioral avoidance,[45,46] and children with more anxious attachment were more strongly influenced by parents' verbal threat information.[45]

Longitudinal studies on the links between parents' verbal communication and children's anxiety have focused on broader characteristics of parental communication, such as negative and positive statements and elaboration.[47] To our knowledge, the longitudinal effects of maternal verbal threat information on the development of childhood fear/anxiety and avoidance of novelty have been studied only in a specific context by Murray and colleagues.[48,49] They compared socially anxious vs control mothers' narratives to their children about school, and measured children's subsequent school representations (via the doll play task) and later functioning. Children of mothers with SAD were more negative/anxious in their school representations. As expected, parents with SAD attributed more threat to school experiences and more vulnerability to their children.[48] Moreover, higher levels of parental threat attributions predicted a higher likelihood of SAD the next term, but only in securely attached children.[49] This latter finding is interesting in that it reveals that a secure attachment style can create more vulnerability to parents' threat attributions. In contrast, studies of cross-sectional links between GAD and insecure attachment in adolescence have revealed that less attachment security is linked to more GAD symptoms/ higher likelihood of GAD diagnosis.[50,51] A longitudinal study revealed bidirectional links between adolescents' perception of their attachment quality with their father and GAD symptoms, whereas only GAD symptoms predicted lower attachment quality with mothers (and not vice versa). Taken together, the findings reveal that insecure attachment to parents may lead to more GAD in adolescents, and vice versa.[52] It is unknown whether differences in findings on the links of anxiety and attachment security are related to the differences in age and anxiety disorder subtype.

Parental behaviors that enhance child anxiety and avoidance

Theoretical accounts of anxiety development[53-55] additionally emphasize the aspects of parental behavior that may contribute to maintenance of fear/anxiety and avoidance in children, for example, by reinforcing avoidant reactions and by limiting children's independence and exposure to stimuli. Parents with GAD, perhaps even more than parents with other anxiety disorders (eg, see Messer and Berdel[56] and Whaly et al[57]) may have a tendency to overcontrol their child's environment because of their generalized fear that something threatening could happen to their child. Hence, they may try to take control over ambiguous situations, which in turn decreases the child's opportunity for exposure, favoring avoidance. Moreover, GAD parents may be overprotective of their children, that is, they may be excessively concerned with the child's safety and be overly cautious with their children, thereby limiting their children's exposure to various situations. In line with theoretical accounts highlighting the potential links of overcontrolling and overprotective parenting styles to child anxiety, meta-analytic evidence reveals a medium-effect (d=0.52 to 0.58) size association between parental overcontrol and child anxiety and a small but significant association between parental overprotection and child anxiety (d=0.12).[58-60]

Similarly, lack of parental reinforcement of exposure or approach to novelty has been implicated in the maintenance of anxious behavior in children. For example, Murray and colleagues showed that socially anxious mothers who fail to encourage approach to strangers promote more avoidance of strangers in their temperamentally fearful infants.[31] Note, however, that replication by Aktar and colleagues revealed a link between more parental encouragement and more child avoidance in infancy and in toddlerhood.[29,41] Thus, parental encouragement of approach could acquire a challenging quality in direct confrontations with novelty. Moreover, the study by Murray and colleagues on the role of maternal narratives about school revealed that SAD mothers were less likely to verbally encourage their children about school.[48,49] Less verbal encouragement from parents predicted more negative school representations and higher levels of internalizing problems in temperamentally fearful children the next term.

Some evidence suggests that parents may reinforce avoidant solutions in their discussions with anxious children: Barrett and colleagues tested 7- to-14-year-old anxious (GAD, SAD, and separation anxiety) children's reactions to ambiguous scenarios involving potentially threatening situations before and after they discussed their reaction with their parents.[61] Results showed that anxious children increased their avoidant strategies after discussion with their parents. Further analysis revealed that parents of anxious children support their child's avoidant strategies.

Nonspecific pathways in the parent-to-child transmission of anxiety: implications for parent-to-child GAD transmission

Evidence reveals that anxious parents can transmit fear/anxiety and avoidance to the offspring by showing nonverbal fear/anxiety signals and by verbally communicating threat information about ambiguous stimuli and situations. Building on this evidence, we propose that elevated anxiety expressions and verbal threat information from GAD parents create multiple learning/modeling opportunities for children to acquire fear/ anxiety and avoidance in a variety of ambiguous situations. These not only include specific contexts addressed in earlier studies (eg, fear of strangers, unknown animals, toy robots), but also extend beyond those (eg, fear of taking a test, not succeeding in life, accidents, illnesses, what will happen to the world). Repeated exposure to parental nonverbal and verbal anxious behavior in children with GAD parents may, alone or in interaction with other vulnerabilities, contribute to GAD development. Broader characteristics of parental communication, such as negative and positive statements or rational versus emotional content await further research in the context of GAD.

On the basis of evidence from specific anxiety disorders, we additionally suggest that parents with GAD may contribute to the maintenance of their children's fearful/anxious and avoidant reactions in the following ways: by showing overcontrolling or overprotective parenting, by reinforcing avoidance, by limiting exposure, and by failing to encourage approach to novelty. In line with this idea, parents with GAD (vs nondiagnosed) were found to be more passive and less encouraging during their interactions with their child in a series of tasks involving playdough, a mysterious box, and social speech.[63] GAD parents also promoted more avoidance of the mysterious box, whereas they did not differ from nondiagnosed parents in modeling of anxiety, in warmth, or in intrusiveness. The question of which specific parenting dimensions may be impaired in specific contexts awaits further investigation in parents with GAD.

Specific pathways in the parent-to-child transmission of GAD: parent-to-child transmission of experiential avoidance and worries

In addition to a generalized form of trait anxiety that is, to some extent, shared with other specific anxiety disorders, worry, a maladaptive verbally rooted coping strategy is a central characteristic of GAD.[24] Worry is what distinguishes GAD from specific forms of anxiety disorders. What differentiates parent-to-child transmission of worry from the nonspecific verbal fear acquisition pathway addressed above is that the latter is triggered by an ambiguous, anxiety-provoking aspect that is present in the environment, whereas worry does not require the threatening aspect to be present now. Worries are directed to the possibility of future, rather than actual occurrences of threat. In this section, we summarize theoretical frameworks addressing the role of worry in the development and maintenance of GAD to shed light on the transmission of worry from GAD parents to children.

Theoretical accounts of GAD conceptualize worry as a cognitive avoidance strategy that inhibits the experience and processing of negative emotional stimuli, and thereby contributes to prolonging the anxiety response.[63-66] Avoidance models propose that worry is generated to cope with future potential threats at a cognitive level.[63,66,67] It prepares for future confrontations with the threat and future reactions at a behavioral level.

Are worries transmitted from GAD parents to the offspring? A study by Pasarelu and colleagues, investigating the associations between mothers' and children's worry and generalized anxiety symptoms in 11- to 17-year-old children revealed that the relation between more worry in mothers and more generalized anxiety symptoms in children was fully mediated by child worries.[68] Therefore, parent-to-child transmission of worries may be an important pathway in the intergenerational transmission of GAD. Thus, parents with GAD may transmit anxiety by communicating to their child their appraisals and worries of potential future threats.

In addition to the direct overlap between worry in parents and children, certain aspects of parenting behavior may feed worries in offspring.[69,70] Muris and colleagues reported that children worried more when they perceived their parents as more rejecting.[70] Higher levels of anxious rearing, as well as insecure attachment styles, seem to predict higher levels of worry in children. Because parenting dimensions have not yet been investigated together with parental worry, it is unclear if the link between these parenting dimensions and child worry remain after accounting for parents' own worries.

Two other individual characteristics, highlighted in the theoretical accounts of GAD are important to shed light on intergenerational transmission of worries: experiential avoidance[71] and intolerance of uncertainty.[72] Defined as the tendency to avoid feeling strong (negative) emotions, experiential avoidance has a central role in theoretical models of GAD. Placed within parenting context, experiential avoidance is the tendency of parents to help their child avoid strong negative emotions like anxiety/fear.[71] Parental experiential avoidance also helps parents avoid their own strong negative emotions triggered by observing their child's experience of anxiety /fear. Parents with GAD may either intervene to control the situation or excessively reassure the child. Excessive parental reassurance may prevent the child from developing autonomous fear coping strategies and, possibly, foster reassurance seeking of the child in future feared situations. Moreover, excessive worry triggered by a child's confrontation with potential threats in daily life, in the case of parental GAD, may mean that parents will be less emotionally available for their children when they are preoccupied with their own worries about their children's fears. In line with this idea, inducing worry in GAD parents of 6-month-old infants decreased parent-to-infant vocalizations, as well as responsiveness to their infant's vocalizations.[73] Note, however, that natural observations of GAD parents' interactions with their infants show that GAD parents are not less sensitive or less than parents without psychopathology[74]; thus, the differences seem to be rather specific to worrying episodes.

Intolerance of uncertainty[72] refers to the discomfort that individuals with GAD experience when confronted with uncertainty/ambiguity in everyday life. The potential dangers embedded in the uncertain situations make GAD individuals avoid these at the behavioral and cognitive levels. Seeing threat in ambiguity due to their own lack of tolerance of uncertainty, and not being able to confront these threats due to their own experiential avoidance, GAD parents are likely to model for their children their own negative cognitions about uncertainty and worry as a coping strategy.

The investigation of the familiality of GAD-relevant characteristics, such as intolerance of uncertainty, experiential avoidance, and worry is still in its infancy; the only available evidence is confined to the study by Pasarelu,[68] which implies that an overlap between parents' and offsprings' worries may account for parent-to-child transmission of GAD. The extent to which the hypothesized overlap in GAD parents' and children's intolerance of uncertainty, experiential avoidance, and worry is accounted for by environmental (vs genetic/dispositional) influences remains to be investigated in future twin studies.

Discussion

This review builds on two defining features of GAD — generalized trait anxiety and worrying — to discuss potential pathways for parent-to-child environmental transmission of GAD ( With respect to acquisition of generalized anxiety, we addressed the potential role of parental verbal and nonverbal anxious/ negative signals on children's acquisition of fear/anxiety and avoidance of ambiguous stimuli in specific contexts. In light of the summarized evidence, we propose that observational and instructional learning of fear/ anxiety and avoidance are two important pathways for transmission that await further research in the context of GAD. We also addressed parental behaviors that may contribute to the maintenance of acquired fearful/ anxious and avoidant/fearful tendencies. To that end, in line with earlier models on child anxiety,[53-55] we propose that parenting behaviors that limit children's chances for exposure to anxiety-provoking stimuli, thereby prolonging fear/anxiety, may contribute to the intergenerational GAD transmission. With respect to specific parent-to-child transmission of worries in GAD, we stressed parents' maladaptive use of worry as a way of coping with high levels of experiential avoidance and intolerance of uncertainty. We propose that GAD parents may encourage child worries by modeling a view of the world as full of potential dangers, but also by promoting worrying as a coping strategy. Below, we briefly address the mechanisms of transmission that may explain child acquisition of fear/anxiety, avoidance, and worry from GAD parents.

The mechanisms explaining parent-to-child transmission of fear/anxiety and avoidance via parental verbal and nonverbal anxiety expressions have been operationalized within the framework of associative learning.[27,43,75] Observational learning of fear/anxiety emerges in development earlier than language and instructional learning and it may operate via the emotional brain systems involved in classical conditioning.[27] Parental nonverbal signals of anxiety during confrontations with novelty act as unconditional stimuli that inherently evoke stress and fear reactions (unconditioned response) that become associated with the novel stimulus and evoke a conditioned fear response.[28,75] Parental verbal communication of anxiety and worry was proposed to activate the mental representation of the ambiguous situation together with threat, establishing and/or strengthening its association with the fear/anxiety response.[43] Additionally, parental enhancement of fear/anxiety and avoidance has been operationalized in terms of operant conditioning processes that reinforce fearful/anxious and avoidant behavior in children and enhance its future likelihood.[19]

The intergenerational transmission of information-processing biases[76] from parents to offspring has been stressed as another potentially important mechanism in parent-to-child transmission of anxiety. Creswell and colleagues propose that anxious parents' vigilance to detect ambiguity/threat in their own, and their child's, environment gives rise to parental anxious behaviors that increase children's vigilance to threat at different levels of information processing (attention, selection, interpretation, and memory bias).[76] Similarly, Field and Lester[77] proposed that each confrontation of the child with ambiguity is a trial in real-life bias training, in which parents guide their children's attention toward threat or safety. In GAD, parents may bias their children's processing of ambiguity by conveying the message that the world is not safe, that uncertainty is intolerable, that strong emotions should be avoided, and that worry and reassurance help to cope with uncertainty, thereby transmitting cognitive styles that characterize GAD.

In line with theoretical accounts and empirical findings on anxiety,[78-82] our findings highlight the importance of considering additional vulnerabilities in the child and in the parent-child relationship as potential modulators of the impact of environmental influences in the study of intergenerational GAD transmission. Notably, studies investigating behavior in everyday interactions in GAD parents have predominantly focused on mothers. As stressed by recent theoretical approaches[82-85] and empirical evidence,[41,42] it is important to include fathers in future studies of GAD transmission. Finally, despite the present exclusive focus on GAD transmission from parents to children, children who — because of a genetic vulnerability — display strong anxiety and worry will become a major source of anxiety and worry for their GAD parents, and as such shape their parenting. The bidirectionality of influences awaits attention in the study of familiality in GAD.

With this paper, we have aimed to provide an overview of potential environmental pathways to intergenerational transmission of GAD to stimulate further research on environmental GAD transmission from parents to children. Our review reveals the paucity of research that specifically focuses on parent-to-child transmission of GAD. Further research into these pathways in family studies that incorporate environmental processes along with genetic and epigenetic influences remains essential before making firm conclusions about which pathways account for intergenerational GAD transmission.