Promoting child mental health has been identified as a key public health issue [1-3]. Most forms of child mental health treatment involve parents, with many efficacious interventions even focusing exclusively on parents, most commonly in the form of parent skills training [4,5]. Parent-based skill training, or parenting education, can take the form of parenting groups, individual treatment, self-help (such as parenting books), and web-based programs. A wealth of evidence suggests that parent skills training programs are effective in reducing child internalizing and externalizing symptoms [6-9]. In addition, parenting education with a prevention focus has been shown to reduce the risk of a wide range of youth emotional and behavioral problems over follow-up periods of up to 20 years [10]. Thus, parent skills training programs are clearly an effective way to treat and prevent child mental health problems.

Despite the demonstrated efficacy of parenting education programs, low parental engagement in these programs has been recognized as a significant barrier to improving child mental health [11]. A systematic review of parental engagement found that 25% of parents who met the criteria for behavioral parent training interventions did not enroll in treatment, and an additional 26% dropped out before the end of treatment, leading to 51% of identified eligible parents not completing treatment [12]. Thus, most families who may stand to benefit from parent training services do not receive a full dose of treatment. Engagement rates in preventive parenting interventions have been found to be similarly low or even lower than those in standard interventions [13,14].

Unfortunately, limited engagement can compromise the ability of parenting education programs to provide desired outcomes for children and families and can temper conclusions drawn from parent skills training research [15]. Moreover, programs with underenrollment are less cost-effective, limiting the effectiveness and disseminability of parenting intervention programs [16]. A host of parent-level, community-level, and programmatic factors have been associated with lower levels of engagement in prevention programs, including socioeconomic disadvantage, lack of social support, single parent status, and minority status [13]; younger parental age and neighborhood disadvantage [11]; parental attributions (eg, external locus of control regarding child behaviors) [17,18]; and limited parental knowledge of efficacious child treatments [19]. A recent study also examined factors that positively influence intent to engage in a parenting intervention in a very large web-based sample (N=6733). Parent behaviors (ie, more coercive parenting, lower consistency, greater use of positive encouragement, and more help-seeking behaviors) and parent cognitions (ie, lower sense of parental self-efficacy, greater psychological distress, and lower perceived quality of parent-child relationship) emerged as significant positive predictors of intent to engage in a parenting program [18].

Piotrowska et al [20] outlined a model of parent engagement with 4 stages: connection, attendance, participation, and enactment. Connection, the earliest stage of engagement, occurs when parents decide to enroll. Importantly, intent to enroll significantly predicts subsequent enrollment and is a moderate-strength predictor of first-session attendance [21,22]. However, despite the underenrollment of parents in prevention and intervention programs to promote child mental health, few studies have experimentally investigated methods to increase program engagement, with only a handful of methods demonstrating efficacy in the early stages of parent engagement. These have included a promotion-focused advertisement [23], a comprehensive “engagement package” (ie, a testimonial flyer, teacher endorsement, and extra calls from group leaders) [24,25], and monetary incentives [26,27].

Overall, few experimental studies have attempted to increase initial engagement. Furthermore, studies have used inconsistent operationalizations of the different aspects of engagement, experimental studies tend to have low methodological quality, and the diversity of methods to enhance engagement makes generalizations difficult [28,29]. Moreover, most initial engagement knowledge comes from intervention research that examined recruitment and enrollment factors post hoc. In these designs, only information from parents who were interested in enrolling is accessible, which precludes identifying the factors associated with parents who choose not to enroll. Taken together, these reviews suggest that there is a need to test novel methods to promote initial parental engagement in preventive interventions.

A separate line of work suggests that including neuroscience explanations can increase subjective credibility and favorability ratings of written information; this phenomenon is known as the “Seductive Allure of Neuroscience Explanations” (SANE) [30,31]. For example, explanations are viewed as more compelling if they include statements such as “brain scans show that...” or “frontal lobe brain circuitry is known to be involved in...” [30,31]. Some researchers have found that this effect is driven by mere conceptual inclusion of neuroscience information [32], whereas others have found this effect only when superfluous neuroscience text and images are included [33]. A review of the neuroimage bias literature concluded that the effects of including these superfluous images may vary according to contextual characteristics [34]. Furthermore, it has been proposed that using optimistic neuroscience explanations that characterize the brain as neuroplastic may enhance the credibility of the information and promote positive social-emotional responses within an intervention [35]. Although it has been suggested that neuroscience explanations may increase engagement with treatment, to our knowledge, this has not been demonstrated experimentally.

In this study, considering the potential utility of preventive parent training, we examined whether a neuroscience-enhanced video presentation about child development and parent skills training enhanced parental intent to enroll more than a program description without neuroscience (standard video). Focusing on the initial stage of engagement, connection [20], we directly compared 2 different methods of advertising a preventive parent skills training program, maximizing the data collected from both interested and uninterested parents. Our model integrates 2 separate lines of work: one on the impact of neuroscience explanations on credibility and the other on parent intervention engagement. We used a framework consistent with the SANE aiming to expand the menu of engagement strategies. Moreover, this study examined the initial stages of engagement in depth, collecting information on self-reported interest in addition to capturing behavioral proxies of intent to enroll.

Beyond the limitations of using self-selected samples of caregivers in initial engagement research, there is a lack of measures of engagement at this earliest stage. In the context of parents already enrolled in a behavioral parent training program, the Parent Motivation Inventory [36] assesses parent desire for child change along with readiness and perceived ability to change parenting behaviors. Other measures have been developed to assess conceptually related constructs, such as the Parental Attitudes Toward Psychological Services Inventory [37] and the Parent Engagement in Evidence-Based Services (PEEBS) questionnaire [38,39], both of which assess parental attitudes toward mental health services. Importantly, these measures assess parental openness to engaging their child in treatment rather than the acceptability of parent-focused training.

In light of this measurement gap and the goal of this study to examine the impact of neuroscience information on parent willingness to engage with parenting education materials, we developed and validated a measure to assess parental intent to enroll, part of the connection stage of engagement [20]. The Parenting Resources Acceptability Measure (PRAM; Multimedia Appendix 1) was created based on a measure of acceptability of positive parenting strategies [40] and on previous work examining consumer preferences for parenting support delivery methods [41]. This measure was developed and evaluated according to a set of established criteria for acceptability measures [42]. Following the development of the 18 items, the measure validation proceeded in 3 phases. First, the measure was subject to an exploratory factor analysis (EFA), and the factors and total scores were correlated with established measures of other relevant constructs. Validity measures were selected based on published measures assessing conceptually related constructs [38,39] along with constructs that have been shown to relate to parent engagement, including child problems [43,44], social support [13], parental locus of control [17], and parental self-efficacy [18]. Next, this measure was used in the experimental study outlined previously. Finally, a subsample of parents was assessed 1 month following the baseline battery administration, and the PRAM was correlated with self-reported engagement in parenting education resources in the previous month. These findings represent the initial validation of this novel measure of the acceptability of parenting education resources.

We hypothesized that the PRAM would demonstrate strong psychometric properties and relate to theoretically relevant measures. We further hypothesized that a neuroscience-enhanced video presentation about child development and parent skills training would enhance parental intent to enroll, as measured by the PRAM, compared with a presentation without neuroscience information. We also hypothesized that this group would display behavior consistent with higher intent to enroll, operationalized as more mouse clicks and time spent on a page of parenting resources, along with a greater likelihood of requesting parenting resources following the presentation.

The baseline Qualtrics survey included a number of questionnaires presented in a randomized order, followed by an experimental manipulation; a page of parenting resources with which parents could interact; and, finally, the postmanipulation repeated measure.

To determine the validity and factor structure of the PRAM (Multimedia Appendix 1), the measure was subjected to an EFA and internal reliability analysis (ie, the Cronbach α). A 3-factor solution was expected, comprising one factor of acceptability (openness), one factor of rated usefulness (interest), and another factor measuring behavioral intent (likelihood).

Pearson product-moment correlations (r) were used to determine the strength of the relationships among PRAM factors, validity measures, and behavioral outcomes, including retrospective reports of engagement at the 1-month follow-up.

Regarding group differences based on manipulation, chi-square tests examined group differences in whether participants responded “yes” to the question, “Would you like to learn more about resources for parents on positive parenting practices?”

To test for differences in engagement between groups immediately following the manipulation within the baseline survey, a negative binomial regression compared the number of total mouse clicks between groups (neuroscience-enhanced video and standard video), and a one-way ANOVA compared the total time spent on the resource page between the 2 groups (neuroscience-enhanced video and standard video). In addition, repeated-measure (RM) ANOVAs assessed differences in acceptability before and after viewing the videos. One evaluated the pre-post differences in total PRAM scores, and 3 additional RM ANOVAs were conducted on the subfactors of this scale based on the results of the EFA.

Finally, univariate RM ANOVAs using all 3 time points of the PRAM were conducted on the subset of participants who completed the 1-month follow-up survey to assess the prospective differences between the standard video and neuroscience-enhanced video groups.

The Florida State University Institutional Review Board exempted this study on February 26, 2020 (reference STUDY00001059).

Chi-square tests and one-way ANOVAs revealed no differences in demographic variables per condition (standard version and neuroscience-enhanced version; P>.05 in all cases). Pre-existing differences between groups on the outcome and validity variables at baseline were ruled out using 2-tailed independent-sample t tests (P>.05). The descriptive statistics for the study participants and baseline measures are reported in Table 1 and Table 2, respectively.

Internal consistency analysis was conducted for all variables. To reduce the total number of statistics, increase the reliability of the results, and simplify interpretation, scales with poor internal consistency (ie, SDQ Internalizing Problems, PLOC Child Control, PLOC Parental Efficacy, and PSOC Interest) were omitted from further correlational analyses. Table 3 presents a bivariate correlation matrix of the important study variables.

After subscale scores were derived for the PRAM based on the EFA, between 1 and 2 low outliers (ie, SD ≥3) were identified for each factor and the total score on both the pre- and postvideo scores and brought to the lower fence. Skewness and kurtosis for each of these scales before and after were within acceptable ranges (<|1.0|).

We created and tested a novel measure to assess the acceptability of resources for parenting education or parent skills training. The measure showed good psychometric properties and related to several theoretically relevant measures. Using this measure, we examined whether, in the context of a brief presentation on parenting education, the inclusion of neuroscience information on child development affected parental intent to enroll in parent skills training. The first hypothesis was partially supported; that is, from before to after the video manipulation, parents in the neuroscience-enhanced video condition scored higher on rated acceptability than parents in the standard video condition on PRAM factor 2 (“Interest in Learning Parenting Strategies”) and on the total PRAM score. By contrast, changes in scores did not significantly differ between conditions on factor 1 or factor 3 (“Acceptability of Parenting Resources” and “Acceptability of Parenting Websites,” respectively). Interestingly, the group differences in rated acceptability from before to after were found to be driven by decreases in rated acceptability in the standard video condition (in contrast to no significant change from before to after in the neuroscience-enhanced video condition). However, when similar analyses were conducted with the subsample of 1-month follow-up completers, there were no significant group differences across all 3 time points. This may indicate that the impact of neuroscience explanations on acceptability is only short-lived. With respect to behavioral measures (ie, requests for resources, number of mouse clicks, and time spent on the resource page), there were no significant differences by condition. However, all behavioral outcomes indicated levels of engagement in the expected direction, with the neuroscience-enhanced video condition showing nonsignificant higher levels of behavioral engagement. We view the results of this study as a first step toward examining the impact of neuroscience-related information on engagement in prevention and treatment approaches.

The PRAM self-report measure was created for this study to fill the measurement gap in assessing the acceptability of parenting resources or training materials. Three factors emerged, roughly divided in terms of media format (ie, acceptability of parenting resources, interest in learning parenting strategies, and acceptability of parenting websites) rather than by facets of acceptability (ie, openness, usefulness, and likelihood). It is possible that the differences among various levels of intent to engage are less important than the ways in which parents consider engaging. For instance, browsing a website for parenting tips requires very little effort compared with other ways of accessing parenting resources and materials. In line with this, at the 1-month follow-up, most respondents reported having accessed parenting resources in the previous month, and the most commonly accessed resource type was web-based information. Previous studies have shown that most parents find evidence-based parenting information to be acceptable and tend to prefer self-administered formats [41,66]. In this community sample, acceptability ratings appeared favorable across factors and the total score; mean ratings were between neither agree nor disagree and agree, skewed toward agree, and “Acceptability of Parenting Websites” had the highest rated acceptability.

Of note, when the 3 items assessing acceptability of parenting websites (factor 3) were omitted, the PRAM retained excellent internal consistency. Thus, it appears that this measure could be administered as a 15-item measure without a meaningful loss of reliability. However, factor 3 showed multiple unique relations with other variables and, thus, may capture an important swath of parents who have lower levels of interest in information found on mainstream websites. In summary, it is recommended that this measure be modified to include or exclude factor 3 depending on the individual study or intervention purposes.

The PRAM displayed a number of strong psychometric properties. Indeed, the measure earned a passing score on each metric of a set of established criteria [42] used to evaluate measures of acceptability, including reliability, structural validity, criterion (predictive) validity, norms, responsiveness, and usability. Specifically, the PRAM earned ratings of excellent on norms (ie, sample size used to establish norms >100), reliability (ie, all Cronbach α values ≥.80), and structural validity (ie, N>100, N>7×the number of items, and an EFA explaining >50% of the variance). It earned a rating of good on usability (ie, instrument length; between 10 and 50 items) and ratings of adequate on criterion validity (ie, medium correlation between the PRAM and another outcome measured in the future) and responsiveness (ie, statistically significant change over time on at least a medium-sized sample; N>50). Importantly, each factor of the PRAM and the total score showed small to moderate–strength positive correlations with behavioral proxies of engagement, including prospective associations with self-reported engagement. Thus, the PRAM has predictive validity as a measure of parental engagement. To further explore the PRAM’s responsiveness and criterion validity, it should be used in prospective studies of established interventions that enhance the acceptability of parent skills training.

In our sample, higher ratings of acceptability were associated with greater parent-reported child maladjustment (internalizing and externalizing problems) in addition to greater difficulty in controlling their child’s behavior. Interestingly, higher reports of parenting dissatisfaction and greater feelings of parenting inefficacy were related to lower acceptability. Although these findings appear to conflict somewhat with those on child behavioral problems, it may be the case that the parental satisfaction and efficacy constructs may better reflect parental stress or psychopathology than actual child behaviors. Future research should test this hypothesis. Furthermore, parental knowledge of effective child treatments, along with the perceived importance of others’ opinions on child-focused treatments (ie, subjective norms), was associated with greater acceptability. Taken together, our results provide evidence that parents of children who display more problems may be more open to help seeking related to parenting education or skill training. Furthermore, our findings outline parent characteristics that may relate to acceptability, and further work is needed to assess whether these traits, including parental sense of efficacy, satisfaction, knowledge of treatments, and subjective norms, are suitable targets for interventions aimed at increasing initial engagement. Finally, perceived social support related only to acceptability of parenting websites; parents who are less socially connected may also have greater mistrust of or less interest in web-based parenting resources touted by experts (eg, the Centers for Disease Control and Prevention) and may benefit from modified engagement methods.

Regarding the experimental manipulation, we found partial support for the hypothesis that self-reported acceptability differed by condition. Specifically, parents who received additional neuroscience information in the video manipulation (neuroscience-enhanced video) did not change their rated interest in learning new parenting strategies from before to after, whereas those in the control condition (standard video) decreased slightly from before to after. Furthermore, at the 1-month follow-up, there were no group differences. It was expected that ratings of acceptability would increase in both conditions, with greater increases in the neuroscience-enhanced video condition. It is possible that parents across both conditions found the video manipulations uncompelling given that no specific parenting intervention or resource was discussed in the video presentations—both videos discussed parent training generally. In addition, given our sample’s skew toward higher educational attainment, it could be that the information presented in the standard video condition was too basic to be of interest, whereas the neuroscience-enhanced video content appealed more to this demographic. It is possible that the pattern of results would differ in a more educationally diverse sample. Despite these unexpected results, this study provides preliminary evidence that the SANE [30,31] can be extended to engage parents—at least initially—in seeking evidence-based parenting resources or education. Interestingly, the only factor of the PRAM that differed between groups was the “Interest in Learning Parenting Strategies” factor and not the 2 factors that included items alluding to specific media formats (ie, books, web-based courses, websites, and local resources). Thus, the SANE effect may enhance interest in general parenting education by increasing beliefs that parents can benefit from parenting education or resources but not with respect to specific media formats. Although additional research is needed to test models of parent engagement with and without neuroscience information, findings suggest that including information about the child brain and the effects of parent training on child brain development may be effective in early-stage parent engagement.

With respect to behavioral outcomes, there were no significant differences in outcomes by condition. However, for all 3 variables (ie, mouse clicks, time spent on resource page, and requests for information), the neuroscience-enhanced video had nonsignificantly higher levels of engagement. It is possible that the effects of neuroscience information on parental intent to engage are small, such that this sample size was not large enough to detect a significant difference. In addition, it is possible that the effects of neuroscience information are more related to perceptions of acceptability than to behavioral outcomes indexing intent to engage. All 3 behavioral outcomes were theorized to capture parental intent to engage. However, of the 3, only requesting additional resources at baseline was positively associated with prospective use of parenting resources at follow-up. Thus, our behavioral measures may actually reflect other processes, such as general interest or arousal during the survey. It is also possible that behavioral outcomes might have differed at other levels of engagement (eg, actual enrollment and quality of engagement with parent resources). Researchers have highlighted the importance of assessing the effects of engagement techniques at multiple levels of engagement [20,67]. Future work should test the SANE effect on the rates of actual enrollment in a parent skills training program.

It bears noting that the manipulation used was weak, consisting of 37 seconds of additional video content pertaining to basic neuroscience. Although this study was a “proof-of-concept” investigation, future work may achieve greater external validity and larger effects by increasing the dose of the SANE effect (ie, infusing neuroscience information throughout the engagement process). In addition, future work could assess whether the addition of neuroscience explanations to parenting education content enhances outcomes in either engagement or child behavioral improvements. Given the dearth of previous research on this topic, it is difficult to estimate this study’s power to detect true effects. A larger sample size may be needed for a fully powered study.

Following the video manipulation, it is possible that participants found the resources presented to be uncompelling and, thus, general levels of behavioral engagement might have been too low to detect differences across conditions. In this study, we were unable to discern which type of resource parents were most likely to engage with while on the resource page (ie, books, web-based programs, websites, and local resources). Future research could examine which types of resources parents are most likely to behaviorally engage with among a menu of parent resource or training options using methods such as eye tracking and advanced mouse tracking.

Emerging evidence has yielded some support for other theory-driven methods of engagement [28,29], including strategies based on the Health Belief Model [68], which emphasizes attitudes and beliefs about health-promoting behaviors, and the theory of planned behavior and reasoned action [64], which links beliefs and attitudes to perceived social pressure and behavioral capacity to perform an action. The manipulation used in this study—based on the SANE—shared an overlap with the Health Belief Model. Future studies could directly compare engagement strategies based on different theoretical models. In addition, the survey collection methodology may have excluded particularly vulnerable groups of caregivers (eg, single parents caring for multiple young children, those unable to complete a web-based survey at home, or those without access to a computer). Finally, this was a predominantly White, high–socioeconomic status, community sample; SANE effects may differ by race or ethnicity or by socioeconomic status. In addition, SANE effects may differ based on the severity of parent problems, child problems, or parent-child interaction problems. However, with this sample, we were able to examine intent to engage in non–treatment-seeking parents. Most initial parent engagement research focuses on treatment-seeking populations, and influences are examined post hoc instead of a priori. This design allowed for the inclusion of parents across a broad range of behavioral intents to engage with parenting resources.

With respect to future directions, it will be important to test the SANE effect, in addition to the psychometric properties and measure invariance of the PRAM, in more nonparents and non–primary caregivers; caregivers who are treatment-seeking; populations of children with mental or behavioral health diagnoses (ie, clinical samples); and various racial or ethnic, cultural, linguistic, and socioeconomic groups. The effect of presenting neuroscience information related to child development may vary across cultures and across parents with different levels of engagement with mental health care. Indeed, the degree to which these findings may be generalized to other populations—for example, to non-English speakers—is unclear. This study’s lack of linguistic and other cultural diversity is a limitation. Future research should also elaborate and expand on the PRAM. Importantly, the PRAM could be used in research on additional stages of engagement beyond intent to engage (eg, actual enrollment, attrition, and implementation of parenting strategies learned). Future work should also examine the associations between PRAM factors and other constructs, such as parent stress or family empowerment [69]. In addition, the PRAM could be modified for use with different groups (eg, treatment-seeking groups, specific diagnostic groups, and parents with records of child maltreatment) or different prevention contexts (eg, web-based parent interventions and group parenting programs). Finally, this study broadly assessed engagement with parenting education resources; future work should examine the specific effects of engagement interventions on the acceptability of certain types or formats of resources provided. For example, it is unknown whether there are different factors that are associated with parental engagement in evidence-based resources versus other types of parenting resources (eg, a parenting community on social media) or in web-based training programs versus self-help books.

This pilot study represents a novel merging of 2 literatures: the SANE and parent engagement in education or training. Extant research on the SANE effect was extended by testing this effect on parents. Moreover, a novel method of parent engagement was tested, with preliminary evidence suggesting that the presentation of parenting education and resources may be more compelling in the short term with the inclusion of simple child brain neuroscience information. The findings have implications for public behavioral health efforts that target parents and may advance the state of parenting prevention science. Researchers should continue to strive toward a better understanding of the factors that drive parental engagement, developing and testing novel methods to enhance engagement and engaging caregivers as active stakeholders in this process. This study is one of a handful of studies to experimentally examine initial engagement; it is possible that a combination of variegated strategies, or simple behavioral “nudges” (eg, inclusion of neuroscience information, style and wording of advertisements, and other yet unidentified enhancements), will ultimately prove instrumental in increasing parent engagement in parent education and training. To the best of our knowledge, this study also represents the first psychometrically validated measure to assess initial parent engagement.