Do Imagined Intergroup Contact Interventions and Metastereotyping Processes Interact to Predict Expectations for Future Inter-Age Interactions?

ABSTRACT

We examined whether imagining intergenerational contact and activating positive aging metastereotypes could enhance the quality of contact young adult participants anticipated having with older persons and attenuate expectations of communicative nonaccommodation by older adults. We also investigated the role of intergenerational trust in these processes. By increasing perceived trustworthiness, imagining intergenerational contact made young adults less prone to expect older people to nonaccommodate (i.e. communicate in a problematic way), compared to a control group. Relative to those in the metastereotype control condition, participants in the positive metastereotype manipulation anticipated less nonaccommodation from older adults.

KEYWORDS:

• Ageism
• imagined contact
• intergenerational relations
• metaperception

Ageism is a perennial problem (Ayalon, 2020; Levy, 2018). Indeed, Abrams et al. (2011) found more than one-third of participants across 28 European nations reported experiencing ageism during the preceding year, a figure far higher than the proportions experiencing sexism and racism. Something that makes ageism especially difficult to overcome is the tendency for society to be highly age segregated. That is, society is structured in ways that impede generational integration (Vanderbeck, 2007), which reinforces distinctions between old and young and limits opportunities for people of different ages to build the sorts of relationships that combat ageism (Hagestad & Uhlenberg, 2006).

Reducing ageism with (imagined) intergroup contact

A review of ageism interventions found that almost two thirds incorporated intergenerational (i.e., intergroup) contact (Burnes et al., 2019), a particularly effective means of reducing prejudice (Levy, 2018; Pettigrew & Tropp, 2006). Originally, contact studies emphasized face-to-face contact (FtFC) between members of different groups (Hewstone & Swart, 2011). However, other kinds of contact are similarly effective at reducing prejudice. The variant of interest in this study is imagined intergroup contact (IIC), which involves the “mental simulation of a social interaction with … members of an outgroup category” (Crisp & Turner, 2009, p. 234).

IIC offers similar benefits to FtFC while circumventing its drawbacks. For example, logistical difficulties of arranging FtFC intergenerational contact in an age-segregated society are obviated by IIC. Moreover, IIC is less likely to elicit intergroup anxiety than is FtFC, and whereas the quality of a FtFC encounter hinges on multiple factors (e.g., the participation of one’s fellow interlocutor), the quality of an imagined interaction “is completely under the control of the person doing the imagining” (Drury et al., 2017, p. 20).

Although intergenerational FtFC has weaker effects than does FtFC in other intergroup contexts (Pettigrew & Tropp, 2006), intergenerational IIC appears to have stronger effects than does IIC in other intergroup settings (Miles & Crisp, 2014) and has been found to produce numerous desirable outcomes. For instance, it reduces anxiety about intergenerational interactions (Birtel & Crisp, 2012), strengthens younger adults’ inclination to interact with older people (Crisp & Husnu, 2011), increases the favorability of young adults’ attitudes toward older persons (Turner & Crisp, 2010), and leads young people to anticipate communicating with older adults in less patronizing ways (Chen et al., 2017).

Optimizing IIC

Mirroring efforts to optimize the design and delivery of FtFC interventions (Jarrott et al., 2021; Martins et al., 2019), researchers are also striving to better understand when and for whom IIC is most powerful. For example, researchers have shown that IIC more effectively reduces prejudice for members of majority groups (Crisp et al., 2010), for people with stronger initial prejudices (West et al., 2017), and for individuals who lack frequent contact with members of the target outgroup (Hoffarth & Hodson, 2016).

Beyond establishing for whom IIC is most effective, researchers have also demonstrated that modifications to how the IIC scenario is presented can enhance its potency. For instance, merely asking people to identify where and when the IIC episode occurs may amplify its benefits (Husnu & Crisp, 2010). Framing IIC episodes as the start of a friendship (Bagci et al., 2018), or asking imaginers to picture the target as a person with whom they have both similarities and differences can also enhance positive effects of IIC (Ioannou et al., 2017). Interestingly, these results parallel findings from the FtFC literature regarding the value of intergroup friendships for reducing prejudice (Brown & Hewstone, 2005) and people’s desire to maintain “optimal distinctiveness” (Brewer, 1991).

In intergenerational contexts, interventions where FtFC is an element of a broader program appear to be more effective than interventions comprising contact alone (Burnes et al., 2019). Based on findings from recent “IIC + x” investigations, we believe there is corresponding potential to optimize IIC by incorporating additional elements.

Igartua et al. (2019) assigned British participants to imagine either interacting with an immigrant or an outdoor scene (serving as a control). Subsequently, participants read a testimonial – ostensibly written by an immigrant – in which the protagonist emphasized either shared social identity with Britons or identification as Pakistani. Results showed improved outgroup attitudes and openness to intergroup contact among participants who engaged in IIC before reading the testimonial (relative to the control group). These outcomes occurred as a function of increasing identification with the protagonist.

In a second study, Ginevra et al. (2021) sought to reduce schoolchildren’s prejudice against peers with sensory, intellectual, or behavioral challenges or disabilities. The researchers compared the efficacy of a cognitive intervention (education about people with such conditions), a behavioral intervention (an episode of IIC where a peer with one of these challenges was the imagined target), and a combined cognitive + behavioral intervention. The combined intervention was more effective than either the cognitive or behavioral intervention alone, perhaps because “the information provided … facilitated the imagined contact with peers with disabilities” (p. 1276).

Something important about the Ginevra et al. (2021) study, we believe, is that the cognitive component of the intervention was administered prior to having participants engage in IIC. Miles and Crisp (2014) posit that IIC may make FtFC more fruitful, but Ginevra et al.’s findings suggest it may be helpful to till the proverbial soil before even sowing the seeds of IIC. In this study, we tested the possibility that this “tilling” might be accomplished by activating positive metastereotypes.

Metastereotypes and intergroup interaction

Metastereotypes comprise the “expectations people have about how members of other groups view their group” (Koudenburg & Gordijn, 2011, p. 221). For instance (and acknowledging that metastereotypes are not inherently negative), Whites (the ingroup) in North America may believe they are stereotyped by Blacks (a specific outgroup) as racially biased (MacInnis & Hodson, 2012). Likewise, young adults (the ingroup) may believe older adults (a specific outgroup) stereotype them as being irresponsible or naïve (Harwood & Williams, 1998). In the present study, we focus on the metastereotypes of young adults (our ingroup); that is, we are interested in the stereotypes that young adults believe older adults have about them.

There is a long history of research linking age stereotypes to communicative behavior in (and communicative expectations for) intergenerational relationships (Hummert, 1994; Ryan et al., 1986). To date, however, relatively little research on communication in intergenerational relationships has been informed by scholarship on metastereotyping (i.e., how members of one age group believe they are stereotyped by members of another age group). This is unfortunate, for Vorauer et al. (1998) contend that metastereotypes shape intergroup conversations even more profoundly than do stereotypes.

Although negative metastereotypes may deter people from engaging in intergroup interaction or lead them to enter intergroup conversations with trepidation (Lammers et al., 2008; Owuamalam et al., 2013), positive metastereotypes can facilitate intergroup relations. Holding positive metastereotypes may, for instance, attenuate negative affective responses to the prospect of intergroup interaction such as anxiety (Matera et al., 2015) and amplify feelings of positive affect – such as trust – toward the outgroup (Vezzali, 2017), thereby weakening the preference to avoid intergroup interaction (Fowler & Gasiorek, 2020).

According to communication accommodation theory, when people interact with others they carry with them “an initial orientation toward the speaker and the encounter based on … interpersonal and intergroup histories” (Gasiorek, 2013, pp. 605–606). We contend that activating positive metastereotypes could modify the “initial orientation” young adults hold toward intergenerational conversations (whether real or imagined) in constructive ways.

Outcome and mediator variables

Ultimately, we believe that the prior activation of positive metastereotypes may enhance young adults’ predisposition to engage in positive IIC with older adult targets with the result that younger people are apt to (a) anticipate higher quality episodes of future FtFC with older adults, and (b) expect older adults to behave less nonaccommodatively in future interactions. Broadly speaking, nonaccommodation refers to communication that is maladjusted relative to at least one interactant’s needs or preferences. For instance, communication that is experienced as patronizing, insensitive, distancing, or unresponsive can all be considered nonaccommodative (Gasiorek, 2016). Consistent with previous work (Pagotto et al., 2012; Turner et al., 2013), we contend that IIC and the activation of positive metastereotypes may function (independently and interactively) to shape young adults’ expectations regarding the quality of future FtFC with older people both directly and indirectly by enhancing intergenerational trust.

At the interpersonal level, trusting a person entails making attributions about that individual’s integrity and honesty. However, it also involves judging that person to be concerned with the welfare of others rather than only themselves (Larzelere & Huston, 1980). This second aspect of trust is particularly relevant at the intergroup level: It suggests that intergroup trust can be conceptualized as the degree to which members of a social outgroup (a) are believed to be interested in the welfare of other groups, and (b) and seek to benefit people belonging to other groups (as well as those belonging to their own group).

Seen in this way, trust may play a vital role in facilitating positive intergenerational relations (Newman & Hatton-Yeo, 2008). Consider, for instance, that one source of intergenerational tension may be the perception that the generations disagree on what is best for society (Gallup Organisation, 2009), with young adults sometimes believing that older people are apt to prioritize and vote for things that benefit themselves rather than things that will benefit younger generations. Indeed, some have argued that the very root of ageism lies in the belief that older adults cannot be trusted to refrain from overconsuming shared resources (e.g., healthcare) or to step aside and make way for younger people when it is their “generational turn” (North & Fiske, 2017).

The present study was designed to assess whether there are main or interaction effects of manipulating IIC and positive metastereotypes on anticipated contact quality and nonaccommodation. It also was designed to test whether these outcomes occur directly, or indirectly via trust. The following hypotheses constitute the model depicted in Figure 1:

Figure 1. Conceptual model.

H1:

Relative to a control condition, people engaging in IIC will experience more intergenerational trust.

H2:

Intergenerational trust will (a) positively predict the anticipated quality of future contact, and (b) negatively predict expected levels of nonaccommodation from older adults.

H3:

As a function of heightened intergroup trust, people engaging in IIC will anticipate higher quality intergenerational encounters and lower levels of nonaccommodation from older adults.

H4:

Engaging in IIC (compared to a control) will exert a stronger direct effect on trust, and stronger total effects on anticipated (a) contact quality and (b) nonaccommodation from older adults among participants in the positive metastereotyping condition.

Method

Participants

Through CloudResearch, we recruited an MTurk sample of people from the United States holding at least a 99% approval rating. Per the processes of the host university, the protocol was evaluated by peer review. Based on this review (and the design characteristics logged in the university’s ethics portal) the study was considered low-risk. As such, it was not required to undergo evaluation by the human ethics committee.

Participants were N = 268 young adults, pre-screened by CloudResearch to be screened to be aged 19–30 (M_age = 24.93; SD = 2.90). A majority identified as female (54.1%); 40.3% identified as male; 5.2% identified as gender-diverse, and one participant declined to report their gender. Most participants self-identified as White (70.5%), 14.2% as Black or African-American, 13.8% as Asian/Asian American, 9.3% as Latino/Hispanic, 2.2% as American Indian or Alaska Native, and 1.1% as other ethnicities. The median level of educational attainment for participants was some college, and the plurality (47.0%) were employed full time.

Experimental procedures

Metastereotype manipulation

Participants were randomly assigned to a positive metastereotype condition or to a control condition. To activate positive metastereotypes, we showed participants written comments about their ingroup ostensibly provided by members of an outgroup. Specifically, participants saw five scanned comments that they were told had been obtained from a previous survey in which older adults described their stereotypes of young adults (see Appendix A for a sample). As in Wakefield et al. (2013), each extract was handwritten in a different style. Below each comment was a typed transcription “in case the image doesn’t display very well … or the handwriting is hard to read.” Participants in the control condition were simply asked to set aside any distractions, informed that they would be answering questions about how people belonging to different generations think about and relate to one another, and then proceeded to the next section (i.e., the treatment or control condition of the imagined contact manipulation).

Imagined contact manipulation

Participants were also randomly assigned to either imagine themselves walking outdoors (the control condition) or experiencing a positive interaction with an intelligent, friendly older woman aged 70–75. We specified that the interaction occurred as “The two of you are stood [sic] in line at the grocery checkout while the cashier has someone do a price check,” and explained that “As you wait, the two of you strike up a conversation.” Participants in the IIC condition gave the older woman a name (that was embedded in task instructions to make the activity more concrete) and described what they imagined the woman looking like.

In both conditions, participants were told they would have one minute to imagine the specified scenario, and informed that when they clicked to the next page in the survey, an “imagination timer” would begin, and that they would not be able to proceed until 60 seconds had passed. After completing the imagination task, participants were asked to provide a written description of the imagined scenario. Participants assigned to the IIC condition were also asked to report how old they imagined the person they were interacting with to be. This allowed us to confirm that the conversation they pictured was one involving an older adult.

Measurement

Intergroup trust

Larzelere and Huston’s (1980) 8-item dyadic trust scale was modified to measure the extent to which participants saw older people as trustworthy (“As a group, elderly people are usually honest and truthful;” 1 = strongly disagree; 7 = strongly agree). The scale showed good reliability (α = .86). For all scales, items were averaged for a composite score (trust: M = 4.38, SD = 0.94).

Anticipated contact quality

Participants estimated the probability (from 0% to 100%) that future conversations with older adults would be pleasant, cooperative, superficial, and insincere (MacInnis & Hodson, 2012). The latter two items were reverse-coded and the scale was reliable (α = .85; M = 60.70, SD = 19.68).

Expectations of older adults’ nonaccommodation

Six items (McCann et al., 2003) tapped expectations young adults had for how older people might communicate with them during a typical conversation (“They would be closed-minded”). The scale used a 5-point Likert format (1 = strongly disagree; 5 = strongly agree) and was reliable (α = .80; M = 2.93, SD = 0.68).

Results

Analytic approach

A path model was run in Mplus 8.5. Imagined contact (1 = imagined contact, 0 = control) and positive metastereotypes (1 = positive metastereotypes, 0 = control) manipulations were the independent variables, trust served as a mediator, and anticipated contact quality and expectations of older adults’ nonaccommodation were positioned as outcomes. An interaction term between imagined contact and positive metastereotypes was created, and modeled as a predictor of trust, anticipated contact quality, and expectations of older adults’ nonaccommodation.

Conditional effects were coded such that indirect, direct and total effects for each metastereotype condition were obtained, with bootstrapped confidence intervals (10,000 samples). Direct effects refer to the strength of unmediated relationships between IIC and the outcomes of interest (i.e., nonaccommodation or anticipated contact quality). Indirect effects refer to the strength of the relationships between IIC and these outcomes via (i.e., mediated by) trust. Total effects refer to the sum of direct and indirect effects, and therefore the overall effect of IIC on the outcome via all possible combinations of paths in the model. The model was fully saturated so there are no fit statistics to report. Results of the model appear in Table 1. Direct, indirect and total effects for IIC by metastereotype condition appear in Table 2.

Table 1. Model path coefficients for imagined contact and positive metastereotype manipulations predicting trust, anticipated contact quality, and anticipated nonaccommodation from older adults.

	Trust (R^2 = .06)		Contact Quality (R^2 = .44)		Nonaccommodation (R^2 = .51)
Predictor	B	p	B	p	B	p
Imagined contact	0.32	0.057	−3.70	0.136	−0.12	0.184
Positive metastereotypes	0.11	0.559	0.52	0.843	−0.21	0.023
Imagined contact x pos. metastereotypes	0.16	0.491	3.26	0.342	0.05	0.677
Trust			14.98	<0.001	−0.44	<0.001

Table 2. Direct, indirect and total effects of imagined contact manipulation by metastereotype condition.

	Contact Quality		Nonaccommodation
	B	p	B	p
Direct effects
Control	−3.70	0.136	−0.12	0.184
Positive MS	−0.44	0.858	−0.07	0.450
Indirect effects (via trust)
Control	4.77	0.059	−0.14	0.055
Positive MS	7.15	0.003	−0.21	0.003
Total effects
Control	1.07	0.758	−0.26	0.017
Positive MS	6.71	0.051	−0.28	0.022
IMM for trust	2.38	0.491	−0.07	0.496

Note: MS = metastereotypes; IMM = Index of moderated mediation.

Focal analyses

H1 received marginal support; relative to the control condition, participants in the IIC condition experienced marginally more intergroup trust (p = .057). H2 was supported; higher levels of trust were associated with higher anticipated contact quality and lower anticipated nonaccommodation from older adults.

H3 was supported for anticipated nonaccommodation. Participants in the IIC condition anticipated lower levels of nonaccommodation than did participants in the control group (see Table 2; all total effects for imagined contact are significant). H3 was not supported for anticipated contact quality; there were not unconditional effects for IIC on this outcome. That is, there were no significant relationships between IIC and anticipated contact quality (directly or via trust) that were not contingent upon (or qualified by) the activation of positive metastereotypes. However, there were marginally significant conditional effects, as predicted in H4.

H4 predicted that the IIC manipulation would exert stronger effects on both anticipated contact quality and anticipated target nonaccommodation for participants in the positive metastereotyping condition. This was marginally supported for contact quality; for participants in the positive metastereotypes condition, IIC had a marginally significant (p = .051) overall effect. However, it did not exert a significant overall effect on participants in the metastereotype control condition. H4 was not supported for anticipated nonaccommodation: Although IIC exerted a positive effect on this outcome, the magnitude and significance of effects did not differ by metastereotype condition. Likewise, H4 was not supported for trust; the interaction between IIC and metastereotype condition predicting trust was not significant.

Discussion

We tested hypothesized relationships between IIC and two measures gauging young adults’ expectations of interactions with older adults: anticipated contact quality, and anticipated nonaccommodation by older adults. Moreover, we sought to determine whether these relationships occurred indirectly via trust, and whether they were moderated by the activation of positive metastereotypes.

Imagined contact

In previous work, Chen et al. (2017) demonstrated that “expectations can be manipulated” (p. 104) by IIC. Specifically, they found that engaging in IIC shaped young imaginers’ expectations for the communication they would use during an intergenerational interaction. Complementing these findings, we found in the present study that engaging in IIC shaped young imaginers’ expectations for the communication they expected older targets to use during these interactions.

Specifically (and acknowledging that there were no effects of IIC on anticipated contact quality), by increasing the perceived trustworthiness of older adults (relative to participants in a control group), engaging in IIC led young adults to anticipate less nonaccommodation from older adults. Such maladapted talk – where older people appear closed-minded, complaining, insensitive, or judgmental – can sour intergenerational relationships (Williams & Giles, 1996) because it signals “disaffiliation, dissimilarity and/or disconfirmation” (Gasiorek, 2016, p. 86).

If younger adults expect interactions with older persons to be superficial and unpleasant, or expect older persons to demonstrate nonaccommodation, it stands to reason that they would find the prospect of interacting with older people unappealing. Furthermore, carrying such expectations into intergenerational interactions might also lead young adults to behave stand-offishly or defensively, thereby eliciting from older interlocutors precisely the superficial or non-accommodative communication that the younger adults anticipated and dreaded. Promisingly, our findings suggest that IIC may help modify the negative expectations which could sabotage intergenerational relationships.

Metastereotypes

The positive metastereotyping manipulation exerted a direct, theoretically sensible effect on anticipated nonaccommodation. Moreover, there was a marginally significant total effect (and a significant indirect effect) of IIC on anticipated contact quality via trust for people in the positive metastereotype condition that was not present for people in the metastereotype control condition). However, our hypothesis that encountering a manipulation intended to seed the belief that older adults viewed younger people positively prior to engaging in IIC would enhance IIC’s effects was not supported; that is, the hypothesized IIC × metastereotype interaction was not significant for any criterion variables (i.e., trust, anticipated contact quality, or anticipated nonaccommodation).

One possibility is that (because the IIC prompt asked participants to imagine themselves having a positive conversation with a woman characterized as friendly and intelligent) there may have been redundancy between the positivity prompted by the metastereotyping manipulation and the (prompted) positivity of the interaction. Alternatively, per the stereotype content model (Cuddy et al., 2005)—our younger participants may already have considered older people (as a social group) to be warm and genial. If this is the case, presenting them with confirmatory “evidence” (in the form of positive metastereotypes) that this geniality was extended to members of their age group specifically may not have been an effective way of shaping how young adults imagined interacting with an older person.

Although it is discouraging that a tool which appeared to have the potential to amplify the benefits of IIC did not function this way, that the metastereotyping and IIC manipulations independently influenced anticipated nonaccommodation suggests these two approaches to combatting prejudice may still complement one another.

Trust

Consistent with previous findings, our results suggested that trust is an important facilitator of intergenerational relationships: Trust predicted both younger adults’ expectation that the overall quality of future contact with older people would be high, and that those older people would not engage in nonaccommodative forms of communication. It was also the conduit through which IIC exerted an overall effect on expectations of nonaccommodation by older adults. Nonetheless, one of the curiosities of contemporary approaches to mediation analyses is that significant indirect effects can be detected even when a constituent path in the mediation model is not significant (Hayes, 2013). This was the case in our study, with IIC emerging only as a marginally significant predictor of trust.

Because IIC has significantly affected trust in previous IIC studies, it may be that differences in context or variations in the IIC manipulation explain the discrepancy. For instance, although Turner et al. (2013) found an effect of IIC on trust when the outgroups specified were asylum seekers and gay persons, we suspect that the levels of distrust held by citizens of a country toward people seeking refuge there (and of non-gay persons toward gay individuals) exceed the level of distrust most younger adults feel toward older adults. Likewise, although Vezzali et al. (2012) detected a large effect of IIC on trust, their study involved young children (for whom, per Miles & Crisp’s, 2014 meta-analysis, IIC appears to be especially potent), and involved three 40-minute intervention sessions rather than the single 60-second IIC exercise used in our study. It is likely that trust could be more effectively cultivated by repeated use of IIC paired with discussion of what was actually imagined. However, any potential gains would likely come at the expense of the ease and affordability of administration that make IIC such an appealing tool for reducing prejudice.

Limitations and directions for future research

A limitation of this study is that our outcome variables captured only participants’ expectations for future intergenerational conversations. Such expectations likely guide how young adults behave during such encounters. Nonetheless, our study cannot tell us whether engaging in IIC before experiencing FtFC actually affects what people actually do during these episodes.

A second limitation is that we focused solely on potential implications of IIC and metastereotyping for relationships between young adults (aged 18–30) and older adults (defined as 70 or older). This precludes drawing conclusions about how IIC might influence the intergenerational relationships of people in their fourth, fifth, sixth, or seventh decades of life. The intergenerational relations literature often gives middle-aged people short-shrift (Fingerman et al., 2004), and future research should include individuals in this age group as both participants and targets of IIC.

Future researchers should also consider methodological trade-offs related to metastereotype manipulations. Although abstract awareness of metastereotypes can influence intergroup relations (Vorauer et al., 1998), the potency of metastereotypes may be greater when someone believes an outgroup member holds stereotypic expectations about them personally because they belong to a particular group (Kamans et al., 2009). Our model was unable to account for the role of metastereotype personalization because our experimental design incorporated a true control group: They could not report on the extent to which activated metastereotypes were personalized because no metastereotypes were activated. In future studies, researchers might consider whether it is more informative to contrast the effects of positive or negative metastereotypes against one another (which permits the incorporation of metastereotype personalization), or against a control (which precludes incorporating metastereotype personalization). Such decisions might be informed by Vezzali’s (2017) finding that there were few differences in outcomes between a negative metastereotype condition and a control condition, which suggests negative metastereotypes are “the default modality at the prospect of meeting an outgroup member” (p. 256).

Conclusion

By enhancing intergroup trust, IIC attenuated young adults’ expectations that older people would engage in nonaccommodative communication. Following previous arguments that IIC may be a useful preliminary to FtFC, our findings suggest that IIC could set the stage for more constructive episodes of actual contact. For example, young adults could be asked to imagine a conversation with a member of the Baby Boomer generation in which the “Boomer” expresses support for policies that benefit younger adults or concern for the difficulties facing today’s young people. Most optimistically, engaging in IIC prior to interacting with an older adult may help set in motion a positive self-fulfilling prophecy, or at least inhibit known negative self-fulfilling prophecies that can stymie intergenerational relationships (Ryan et al., 1986). We also found some evidence that activating positive metastereotypes might (independently) achieve a similar end, indicating the potential for additive benefits of IIC and positive metastereotyping processes to improve intergenerational relations.

Acknowledgments

Raw data can be obtained at https://osf.io/khbj6/?view_only=88f14aad5e314d758bade0bbafca9e08

Disclosure statement

No potential conflict of interest was reported by the authors.

Supplementary material

Supplemental data for this article can be accessed online at https://doi.org/10.1080/15350770.2023.2208575.

Additional information

Funding

The study was deemed low-risk (Ethics notification number: 4000024241) and was not required to be submitted for full IRB review.