Abstract
Importance:
Sensory seeking (SS) is a behavior in which individuals actively pursue intense sensory experiences. It is commonly described in autistic children and those with other neurodevelopmental differences, as well as in typically developing individuals. Despite its prevalence, there is controversary regarding the underpinnings of this behavior.
Objective:
To determine whether, as proposed in Dunn’s model, SS is related to poor sensory registration in typical adolescents and adults.
Design:
A correlational study examining relationships among factors in Dunn’s model.
Setting:
Schools, universities, occupational centers, and private organizations for adolescents and adults.
Participants:
Typically developing, Spanish-speaking adolescents and adults (>11 yr, n = 787).
Method:
Pearson correlation coefficients were used to examine relationships between SS and poor sensory registration on the Spanish version of Adolescent/Adult Sensory Profile.
Results:
A nonsignificant relationship (r = .004) was found between SS and poor sensory registration. Other sensory reactivity patterns showed some positive relationships.
Conclusions and Relevance:
SS and poor sensory registration pattern did not show strong relationships as suggested by Dunn’s model in this sample. These data suggest that alternate views of the underlying mechanisms of SS may be needed.
Plain-Language Summary
Sensory seeking has been linked to sensory reactivity problems but also to other problems, such as poor praxis and poor sensory perception. This study presents new data showing an absent or weak relationship between sensory seeking and poor sensory registration patterns and proposes alternate models of viewing sensory reactivity patterns and sensory seeking behavior.
This study found an absent or weak relationship between sensory seeking and poor sensory registration patterns and proposes alternate models of viewing sensory reactivity patterns and sensory seeking behavior.
Ayres Sensory Integration® (ASI) theory (Ayres, 1964) describes the relationship between sensation and behavior and includes a subtype of sensory integration referred to as sensory reactivity (SR). In Ayres’s model, SR includes hyper- and hyporeactivity (originally referred to as over- or underresponsiveness, respectively) and was initially identified among autistic children (Ayres & Tickle, 1980). Later, SR was added to the diagnostic criteria for autism in the Diagnostic and Statistical Manual (5th ed.; American Psychiatric Association, 2013), where it was associated with restricted and repetitive behaviors, specifically unusual interest in the sensory aspects of the environment.
Following Ayres, Dunn (1999) proposed a model of sensory processing that focuses mainly on SR and its behavioral consequences. This model describes four SR patterns, including sensory seeking (SS). SS is associated with high sensory thresholds and active self-regulating strategies. According to Dunn (1999) and Brown and Dunn (2002), high sensory thresholds require greater or more intense stimuli to activate neurons; thus, individuals with SS engage in behaviors related to active pursuit of sensation.
Dunn (1999) also hypothesized that children with another SR pattern, poor sensory registration, have high sensory thresholds and inadequate neural activation, thus establishing a hypothesized relationship between SS and poor sensory registration. Consistent with Dunn’s theory, Little and colleagues (Dean et al., 2018; Little et al., 2017) viewed SS as a high-threshold, active self-regulation pattern. In their studies with typically developing individuals and individuals with developmental conditions, SS emerged as one of the distinct sensory subtypes of SR. They showed that SS was associated with challenges in self-regulation, attention shifting, and behavioral flexibility and had an impact on participation in daily activities. Little and colleagues (Dean et al., 2018; Little et al., 2017) maintain that SS is related to high sensory thresholds but also recognize that it may have other potential explanations.
Specifically in relation to autism, Baranek et al. (2006) found that SS is not simply a SR issue but a separate behavioral phenotype in autism that is associated with poorer self-regulation, higher levels of repetitive behavior, and greater challenges in attention and social engagement among autistic persons (Baranek et al., 2006, 2013). Although Baranek et al.’s (2013) work suggests that SS may be a strategy to compensate for low sensory registration, they also acknowledge that it may be related to poor sensory perception or used to generate movement ideas when praxis is inefficient. Similarly, Lane and Schaaf (2010) and Bundy and Lane (2019) acknowledge that SS may reflect neural inefficiencies in perception and praxis as well as difficulty with arousal modulation. These authors suggest that SS does not have a single cause or mechanism but may arise from multiple factors.
Furthermore, historically, SS was described as a behavior associated with individual characteristics or traits of typically developing individuals. For example, Zuckerman (1994) proposed that SS reflects a stable trait characterized by a preference for novel intense experiences and high levels of stimulation, whereas Zentall and Zentall (1983) proposed the optimal stimulation hypothesis, describing SS as a compensatory strategy to maintain optimal arousal. More recent evidence suggests that SS is associated with emotional outcomes, including depressive symptoms and resilience, indicating that it may serve multiple regulatory and adaptive functions beyond simple sensory responsiveness (Dean et al., 2018).
Thus, SS is a behavior that may reflect an SR issue, poor perception, or inadequate praxis or, conversely, a reflection of an individual personality characteristic or trait. Because it is beyond the scope of this study to examine every explanation for SS, this study focuses on the idea that SS is related to a high sensory threshold and poor sensory registration. These findings will add clarity to the mechanisms underlying SS and can guide intervention accordingly. For example, if SS is related to poor sensory registration, interventions may be targeted to enhance sensory activation, whereas if SS and poor sensory registration are not associated, intervention strategies should include other options.
Method
We conducted a correlational study to examine the relationship between the SS and poor sensory registration patterns in Dunn’s sensory processing model using the Adolescent/Adult Sensory Profile (AASP) in the Spanish population.
This study was approved by the Clinical Research Ethics Committee of Galicia. Data collection always preserved participants’ anonymity. All study participants signed the corresponding informed consent form. In the case of minors, their parents or legal guardians signed the consent forms.
Participants
Data from a prior data set used to determine the normative values of the AASP for typically developing adolescent and adults (N = 787) in Spain were used (Gándara-Gafo et al., 2019). Table 1 shows the sample’s sociodemographic characteristics. Inclusion criteria for participants were as follow: older than age 11 yr, Spanish nationality, resident in Spain, signed the informed consent form, and answered the Spanish questionnaire. Individuals diagnosed with depression or chronic disease, such as diabetes, cancer, neurological disease, autism, or attention deficit hyperactivity disorder, or cognitive disabilities, or taking some kind of medication that might affect the nervous system were excluded. To obtain normative values for the AASP in Spain, we contacted different colleges or schools, universities, occupational centers for individuals older than age 65 yr and private organizations by email, and the principal investigator (Berta Gándara-Gafo) visited institutions interested in the study to recruit participants. Participants completed the questionnaire in person.
Sociodemographic Characteristics (N = 787)
Instrument
The AASP (Brown & Dunn, 2002) is one of the few tools available to gather SR data in individuals older than age 14 yr. This tool, consisting of 60 items, is designed to provide information about SR and is used to plan individual intervention on the basis of individual sensory preferences. Items are grouped by sensory modality (taste and smell, movement, visual, touch, body position, auditory), and the results are categorized into one of four SR patterns, namely, poor sensory registration, SS, sensory avoidance, and sensory sensitivity (Dunn, 1997). The AASP (Brown & Dunn, 2002) shows adequate reliability and validity. Internal reliability analysis (Cronbach’s α) is reported to be between .60 and .78 per pattern, and validity was supported by a skin conductance test with a small sample (n = 20) who showed significant differences in responsivity across the four patterns (Brown et al., 2001). Findings showed that each sensory category showed a different physiological response and supported the distinct constructs in Dunn’s model (Brown et al., 2001).
The culturally adapted Spanish version of the AASP (Gándara-Gafo, Santos-del Riego, Viana-Moldes, et al., 2019) has strong test–retest reliability, with values higher than .9 for all the sensory processing patterns. Its internal reliability (Cronbach’s α) was between .69 and .73 per pattern, and it clearly differentiated between individuals with typical development (n = 52) and individuals with schizophrenia disorder (n = 18), indicating discriminant validity (Gándara-Gafo et al., 2023). The Spanish cultural adaptation also contains reference values (N = 787) for individuals with typical development in Spain (Gándara-Gafo et al., 2019). Score ranges for “similar to most people” in the Spanish population, by sensory pattern and age group (11–17 yr, 18–64 yr, and >65 yr), are as follows: poor sensory registration, 24 to 37, 23 to 37, and 26 to 44, respectively; SS, 41 to 57, 41 to 56, and 34 to 53, respectively; sensory avoidance, 29 to 45, 28 to 46, and 32 to 52, respectively; and sensory sensitivity, 27 to 40, 28 to 43, and 32 to 51, respectively. See Gándara-Gafo et al. (2019) for more information regarding means, standard deviations, and ranges in the Spanish population.
Data Analysis
Pearson correlations were used to examine the relationship between the sensory processing patterns in the total sample (N = 787) as well as by age group: adolescents (11–17 yr), adults (18–64 yr), and older adults (<65 yr). Pearson’s coefficient gives a variation ranging between −1 and 1, where 0 is the absence of linear relationship between variables, and −1 or 1 is a perfect negative or positive linear relationship, respectively (Green et al., 2000). Values between 0 and .10 represent an absence of correlation; between .10 and .29, a weak correlation; between .30 and .49, a moderate correlation; and greater than .5, a strong correlation (Lalinde et al., 2018). IBM SPSS Statistics (Version 19.0) software was used for data analysis.
Results
As shown in Table 2, no relationship was found between the poor sensory registration and SS patterns (r = .004) in the total sample (N = 787). Similarly, in the age-grouped samples, no relationship between poor sensory registration and SS patterns was found in adults and older adults (r = .020 and r = .090, respectively). However, a weak correlation was found for adolescents (r = .162, p < .05).
Pearson′s Correlation and Bilateral Significance for the Entire Sample (N = 787) and by Age Group
*p < .05. **p < .01. ***p < .001.
The analysis of relationships among the other patterns found positive correlations between the sensory sensitivity and sensory avoidance patterns (r = .631, p < .01), between the poor sensory registration and sensory sensitivity patterns (r = .583, p < .01), and between the poor sensory registration and sensory avoidance patterns (r = .436, p < .001). Absent or weak correlations were found between sensory seeking and other patterns.
Discussion
This study contributes to the ongoing discourse surrounding the mechanisms underlying SS behavior, using a sample of adolescents and adults with typical development. Despite the proposed relationship between SS and poor sensory registration patterns, both of which are linked to high sensory thresholds, our results revealed no significant relationship between these two patterns. This result challenges the assumption that SS is inherently linked to poor sensory registration and suggests the need to reevaluate the underlying mechanisms of SS.
The absence of a strong relationship between SS and poor sensory registration aligns with prior research across developmental stages and diagnostic groups. Studies by Ben-Sasson et al. (2007) with individuals with autism and by Engel-Yeger (2012) and Gándara-Gafo (2016) with typically developing individuals, reported that poor registration was more often associated with sensitivity and avoidance behaviors rather than active seeking. These converging findings suggest that SS may not be a direct consequence of high sensory threshold or inadequate neural activation, as originally proposed by Dunn (1999), but rather a more complex behavioral phenomenon with multifactorial explanations.
Zentall and Zentall (1983) posited that SS may serve as a compensatory strategy to maintain optimal arousal, and Zuckerman (1994) framed it as a personality trait in typically developing individuals. Dunn (2001) herself acknowledged that temperament may influence SS behaviors. More recently, Bundy and Lane (2019) and Schaaf and Mailloux (2015) have argued that SS may reflect deficits in sensory perception or praxis, as well as sensory reactivity and modulation issues.
When considering the potential influence of age on sensory processing patterns, we did not observe a significant age tendency in this study; interestingly, however, we did find a weak association between SS and poor sensory registration in the adolescent sample only. This finding may be partly explained by developmental changes in sensory processing patterns across the lifespan, because sensory-seeking behaviors tend to decrease with age. Engel-Yeger and Rosenblum (2021) reported that 25% of the typically developing adults in their sample, who were younger than age 65 yr, showed a reduced tendency to seek sensory input. These age-related differences suggest that sensory processing patterns evolve over time and may shape how individuals interact with their environment, potentially restricting participation in daily activities and meaningful occupations.
Our results also revealed an age-related divergence in the relationship between SS and sensory sensitivity. Although adolescents and older adults exhibited positive correlations (r = .211 and r = .210, respectively), adults demonstrated a negative correlation (r = −.22). This directional shift could be linked to the dynamic nature of sensory processing and self-regulation across the lifespan and warrants further examination. The negative correlation could potentially be driven by the use of cognitive strategies for adaptive regulation in adulthood, reflecting the development and use of top-down cognitive control among adults (Rodríguez & Kross, 2023).
In terms of relationships among other sensory processing patterns in Dunn’s model, we found a strong relationship between sensory sensitivity and avoidance (r = .631, p < .01). This finding reinforces Dunn’s conceptualization of these patterns as expressions of low sensory thresholds and aligns with descriptions of hyperreactivity (Bundy & Lane, 2019). We also found moderate correlations between poor registration and both sensitivity (r = .583, p < .01) and avoidance (r = .436, p < .01). These findings are surprising, given that poor registration is theoretically linked to hyporeactivity or poor perception of sensory input. These results could suggest that individuals may exhibit mixed reactivity profiles.
This complexity echoes findings by Boyd et al. (2010), who reported moderate correlations between hypo- and hyperreactivity for children with autism. Such overlap may reflect the dynamic nature of sensory processing across contexts and modalities or may indicate limitations in the current nosology of SR. The emerging bidimensional model proposed by Bundy and Lane (2019)—which conceptualizes sensory reactivity along hypo- and hyperreactive axes—may offer a more parsimonious framework for understanding these patterns.
Taken together, these findings suggest that SS may not be best understood as a subtype of SR but rather as a behavioral expression influenced by multiple factors such as SR, perceptual, motor, and possibly temperamental factors. This reconceptualization has important implications for occupational therapy practice. Clinicians should be cautious in attributing SS behaviors solely to a high sensory threshold or poor registration and instead consider broader assessments that include sensory perception, praxis, and contextual factors. Future research should investigate the role of sensory perception, motor planning, and temperament in shaping these behaviors and examine how these factors interact across developmental stages and diagnostic groups. Furthermore, understanding age-related trends in SS is an important consideration, and we hope to address this in future research.
Although the findings from this study add valuable information to the literature regarding SS, several limitations must be acknowledged when applying these findings. First, because the AASP aggregates SR across modalities, it may potentially obscure modality-specific patterns (e.g., low registration in one modality and seeking in another). Future research should incorporate modality-level analyses. Second, given that the samples for this study were typically developing adults and adolescents, there was likely restricted score variability. This may have attenuated detectable associations and should be considered when interpreting the findings. It is important to acknowledge that the sample in this study consisted of typically developing individuals, which may not accurately reflect sensory processing differences among autistic individuals or those with other neurodevelopmental conditions.
Implications for Occupational Therapy Practice
The findings of this study have the following implications for occupational therapy practice: ▪ The absence of correlation between the poor registration and seeking patterns suggests that clinicians should consider alternate explanations beyond poor registration. ▪ Clinicians may want to consider a two-dimensional model of dysfunction in SR (hyper- and hyporeactivity) rather than a four-dimensional one.
Conclusions
This study’s results call into question the idea that SS is a SR pattern related to poor sensory registration. Furthermore, it supports the idea that the challenges in SR may be two-dimensional: (1) hyporeactivity, referred to as poor registration in Dunn’s model of sensory processing, and (2) hyperreactivity, referred to as sensitivity and avoidance in Dunn’s model. Future studies are needed to further explore the underlying mechanisms of what are often termed SS patterns or behavior.
Footnotes
Acknowledgments
Thanks to the participants who made this study possible. Research was carried out by the Unidad Integra-Saúde (University of A Coruña).