Dr. Who? The Effects of Familiarity, Gender, and an Academic Title on the Perception of Scientists Online

Distinguishing between facts and opinions can be challenging - especially on social media. When scrolling through nearly every common social media platform’s feed the information accessible for users to evaluate senders’ trustworthiness are mostly limited to a profile picture and username. Hence, we focus on these limited cues that are directly visible and derivable by the recipients. We will analyze the degree to which the cues gender, familiarity of the sender, and an academic title (as an optional part of the username) influence scientists’ perception as trustworthy. From these analyses, we will derive implications for science communicators’ self-presentation online. Grounding on the perspective of public online engagement with science (Taddicken & Krämer, 2021) the question arises whether a direct affective relationship of trust with a sender is significant for impactfully communicating science and promoting engagement. The closely related construct of source credibility (McCroskey & Teven, 1999) provides reasons to suggest that an affective component can influence users’ assessment of trustworthiness. Here, affective trust is based on emotional ties of (interpersonal) relationships and the maintenance of positive expectations concerning the emotional investment made in the relationship (Lewis & Weigert, 1985). We will explore whether the perception of an affective and personally trusting relationship is associated with the general trustworthiness of scientists online. The focus here is on the recipients’ perceived goodwill of the senders towards them and not on the anticipated benevolence of the scientists towards the public. The option to engage with a scientist directly on social media without any institutional or journalistic barrier gives recipients and senders at least the opportunity to build a trusting relation – even if it is one-sided. Therefore, it is necessary to enlighten if and how this type of trust has an impact on communicating science and research online. According to research on trust in science and scientists, a trustworthy source and sender needs to creates an impression of expertise, integrity as well as benevolence (Hendriks et al., 2015, 2016). Since epistemic trustworthiness refers to the trust placed in others' knowledge (Wilholt, 2013), it describes qualities of a sender against the background of a difference in anticipated personal dependence plus need for trust concerning the sender’s characteristics. Given the wealth of scientific knowledge, not every type of information can be gained individually by learning or by recipients’ firsthand experience. In consequence, recipients need to rely on trustworthy sources (Hendriks et al., 2015, 2016). Up to now, little research has been carried out concerning the cues that are associated with an impression of epistemic trustworthiness recipients based their judgments on. Especially from the perspective of science communication it is crucial to highlight a sender’s expertise since its needed to promote trust (Hendriks et al., 2015). This raises the question of whether a reference to a successfully completed academic education is enough for a recipient to evaluate the content of a scientists’ message as credible, or whether other factors concerning the sender are decisive or not. Previous studies indicate that conveying expertise cues has an influence on the perception of the sender’s credibility and whether recipients trust the information shared by them or not (Hendriks et al., 2015; Jucks & Thon, 2017; Lin & Spence, 2018; Nadarevic et al., 2020; Pjesivac et al., 2018; Winter & Krämer, 2014). One way of conveying expertise online is to refer to an academic education - i.e., by a visible academic title. We expect expertise cues, like in this case presenting an academic title, will serve as a “rational reason” to trust and will thereby be associated with cognitive trust. It approaches more rational aspects and binds trust to rather objective criteria like professionalism (Lewis & Weigert, 1985; McAllister, 1995). Nonetheless, when it comes to the formation of trust, more factors are decisive. It should be emphasized that epistemic trust, especially with the dimension of benevolence, encompasses senders who do not know the recipients personally and are therefore more or less “strangers on the internet”. Since familiarity can affect trustworthiness via various associated phenomena, two likely types are addressed here. Frist are benefits due to an increased fluency of processing. Recent evidence suggests that repeatedly seen messages are perceived as more credible, believable, and trustworthy (Luo et al., 2022; Nadarevic et al., 2020; Pennycook et al., n.d.; Pennycook & Rand, 2020; Stump et al., 2022), which can be explained, among other factors, by the "truth effect". It is one of several well-known phenomena, such as the mere exposure effect, which promises positive outcomes through repetition. Here, the general rule of thumb is that the better known a stimulus is, the easier it is to process (see Dechêne et al., 2010 for review). So, it can be assumed that a similar effect could also be found regarding the perception of messages’ senders and scientists, respectively. Notwithstanding, a second explanatory approach in the context of trust is also conceivable. Trust is formed in anticipation of the counterpart’s future behavioral expectations, whereby trust is created ex post facto and legitimized when fulfilled (Luhmann, 2014). If a sender is familiar to the recipient (e.g.: has been seen beforehand) and this previous interaction was not unpleasant or connected to a betray of epistemic trust (e.g.: due to information subsequently identified as false) then no contraindication to trust should exist. We assume that the familiarity of a sender can be one factor influencing trust since recipients might fall back on a "pre-validation". This refers to prior firsthand experiences of the recipients regarding the sender. The existence of preexisting attitudes should only be present and impactful if the sender is familiar. This might sound obvious, however at this point research on science communication online is missing out that notion. If the pre-established attitudes do not miss the benefit of a doubt, this should result in a favorable impression of a sender and lead to an increase in the senders’ trustworthiness on an epistemic level or the other way around. Further, asking if there are any other cues that lead to recipients to assign epistemic trustworthiness to sender, we plan to explore if the senders’ gender affects the recipients’ perception as well. Gender is used as a category to estimate judgements with and provides a starting point to evaluate – for instance based on stereotypes. This automagical judgement based on gender, might also apply on scientists communicating their research as it can influence readers’ assessment as credible (Winter & Krämer, 2014). When conveying information males often were found to be perceived as more credible compared to women (Armstrong & McAdams, 2009; Embacher et al., 2018; Knobloch-Westerwick et al., 2013). Though, more recent studies show opposite results concerning the general credibility assessment online (Bhandari et al., 2021) and indicate a change in stereotypical gender roles in terms for scientists for a younger generation (Bigham et al., 2019). As a sender’s gender is conveniently conveyed in a profile picture in users’ social media feeds, we further aim to investigate its impact on the scientists’ trustworthiness. Lastly, we will include three constructs to control the results: The (general) trust in science as well as participants beliefs about the scientific consensus regarding COVID-19 and their level of perceived uncertainty.