Psychological Insights for Understanding COVID-19 and Media and Technology

Ciarán Mc Mahon

• Publisher: Routledge
• Year: 2020
• eBook ISBN: 9781000337037
• Topic: Medicine
• Subtopic: Diseases & Allergies

In the Psychological Insights for Understanding COVID-19 series, international experts introduce important themes in psychological science that engage with people's unprecedented experience of the pandemic, drawing together chapters as they originally appeared before COVID-19 descended on the world.

This book explores how COVID-19 has impacted our relationship with media and technology, and chapters examine a range of topics including fake news, social media, conspiracy theories, belonging, online emotional lives and relationship formation, and identity. It shows the benefits media and technology can have in relation to coping with crises and navigating challenging situations, whilst also examining the potential pitfalls that emerge due to our increasing reliance on them. In a world where the cyberpsychological space is constantly developing, this volume exposes the complexities surrounding the interaction of human psychology with media and technology, and reflects on what this might look like in the future.

Featuring theory and research on key topics germane to the global pandemic, the Psychological Insights for Understanding COVID-19 series offers thought-provoking reading for professionals, students, academics and policy makers concerned with the psychological consequences of COVID-19 for individuals, families and society.

1
WHEN (FAKE) NEWS FEELS TRUE

Intuitions of truth and the acceptance and correction of misinformation

Norbert Schwarz and Madeline Jalbert

An analysis of 2.8 million episodes of news sharing on Twitter found that 59% of the news items were shared without having been opened (Gabielkov, Ramachandran, Chaintreau, & Legout, 2016). Apparently, six out of ten readers found the headline compelling enough to share the piece without reading it. In this chapter, we review what makes a message “feel” true, even before we have considered its content in any detail. We first discuss the basic psychological processes involved in assessing the truth of a message and illustrate them with select experiments. Subsequently, we address the implications of these processes for information sharing on social media and the correction of misinformation.

Evaluating truth

While retweeting something without reading it may strike many readers as surprising and irresponsible, it is not distinctly different from how we communicate in everyday life. In daily conversations, we proceed on the tacit assumption that the speaker is a cooperative communicator whose contributions are relevant to the ongoing conversation, truthful, informative, and clear (Grice, 1975; Sperber & Wilson, 1986). Unless we have reason to doubt that the speaker observes these tacit rules of conversational conduct, we accept the content of the utterance without much questioning and treat it as part of the common ground of the conversation. These conversational processes contribute to many errors in human judgment (for reviews, see Schwarz, 1994, 1996). Some research even suggests that comprehension of a statement requires at least temporary acceptance of its truth (Gilbert, 1991) before it can be checked against relevant evidence.

While suspension of belief is possible (Hasson, Simmons, & Todorov, 2005; Schul, Mayo, & Burnstein, 2008), it requires implausibility of the message or distrust at the time it is received. Hence, the deck is usually stacked in favor of accepting information rather than rejecting it, provided there are no salient markers that call the speaker’s cooperativeness into question. Going beyond the default of information acceptance requires motivation and cognitive resources, which we are most likely to invest when the topic is important to us and there are few competing demands and distractions. In the absence of these conditions, information is likely to be accepted – and sometimes passed on – without much scrutiny.

When people do evaluate whether information is likely to be true, they typically consider some (but rarely all) of the five criteria shown in Table 5.1 (Schwarz, 2015). Is the claim compatible with other things they know? Is it internally consistent and coherent? Does it come from a trustworthy source? Do other people agree with it? Is there much evidence to support it? Each of these criteria is sensible and does, indeed, bear on the likely truth of a message. These criteria can be assessed by considering relevant knowledge, which is a relatively slow and effortful process and may require extensive information search. The same criteria can also be assessed by relying on one’s intuitive response, which is faster and less taxing. When the initial intuitive response suggests that something may be wrong, people are likely to turn to the more effortful analysis, provided time and circumstances allow for it. This makes initial intuitive assessments of truth a key gatekeeper for whether people will further engage with the message using a critical eye or just nod along in agreement. These assumptions are compatible with a long history of research in social (e.g., Petty & Cacioppo, 1986) and cognitive (e.g., Kahneman, 2011; Stanovich, 1999) psychology, where the slow and effortful strategy is often referred to as “analytic”, “systematic”, or “system 2” processing and the fast and intuitive strategy as “intuitive”, “heuristic”, or “system 1” processing.

TABLE 5.1 Truth criteria

Criterion	Analytic evaluation	Intuitive evaluation
Compatibility: Is it compatible with other things I know?	Is this compatible with knowledge retrieved from memory or obtained from trusted sources?	Does this make me stumble or does it flow smoothly?
Coherence: Is it internally coherent?	Do the elements fit together in a logical way? Do the conclusions follow from what is presented?	Does this make me stumble or does it flow smoothly?
Credibility: Does it come from a credible source?	Does the source have the relevant expertise? Does the source have a vested interest? Is the source trustworthy?	Does the source feel familiar and trustworthy?
Consensus: Do other people believe it?	What do my friends say? What do the opinion polls say?	Does it feel familiar?
Evidence: Is there supporting evidence?	Is there supportive evidence in peer-reviewed scientific articles or credible news reports? Do I remember relevant evidence?	Does some evidence easily come to mind?

Key to intuitive assessments of truth is the ease with which the message can be processed. For example, when something is incompatible with other things we know or the story we are told is incoherent, we stumble and backtrack to make sure we understood it correctly (Johnson-Laird, 2012; Winkielman, Huber, Kavanagh, & Schwarz, 2012). This makes the subjective experience of ease of processing, often referred to as processing fluency, a (fallible) indicator of whether the message may have a problem that needs closer attention. Similar considerations apply to the other truth criteria, as discussed later in the chapter. Throughout, difficult processing marks the message for closer scrutiny, whereas easy processing favors message acceptance.

If ease or difficulty of processing was solely determined by attributes substantively associated with whether a message is likely to be true, relying on one’s processing experience would not pose a major problem. However, messages can be easy or difficult to process for many reasons – reading may be slow because the message is incoherent (a relevant criterion) or because the print font is hard to read (which is unrelated to truth). Because people are more sensitive to their subjective experiences than to the source of those experiences (Schwarz, 2012), many incidental influences that have no bearing on the substance of the message can influence its perceived truth. We discuss these incidental influences and their role in media consumption after reviewing the five dominant truth criteria. As will become apparent, when thoughts flow smoothly, people are likely to agree without much critical analysis (see also Oyserman & Dawson, this volume).

The “big five” of truth judgment: analytic and intuitive processes

A claim is more likely to be accepted as true when it is compatible with other things one knows than when it is at odds with other knowledge. Compatibility can be assessed analytically by checking the information against one’s knowledge, which requires motivation and time (Petty & Cacioppo, 1986). A less demanding indicator is provided by one’s metacognitive experiences and affective responses. When something is inconsistent with existing beliefs, people tend to stumble – they take longer to read it, and have trouble processing it (e.g., Taber & Lodge, 2006; Winkielman et al., 2012). Moreover, information that is inconsistent with one’s beliefs produces a negative affective response, as shown in research on cognitive consistency (Festinger, 1957; Gawronski & Strack, 2012). Accordingly, one’s processing experiences and affective responses can serve as (fallible) indicators of whether a proposition is consistent with other things one believes.

A given claim is also more likely to be accepted as true when it fits a broader story that lends coherence to its individual elements, as observed in research on mental models (for a review, see Johnson-Laird, 2012) and analyses of jury decision making (Pennington & Hastie, 1993). Coherence can be determined through a systematic analysis of the relationships between different pieces of declarative information. Alternatively, it can be assessed by attending to one’s processing experience: coherent stories are easier to process than stories with internal contradictions (Johnson-Laird, 2012), which makes ease of processing a (fallible) indicator of coherence. Indeed, people draw on their fluency experience when they evaluate how well things “go together” (Topolinski, 2012), as observed in judgments of semantic coherence (Topolinski & Strack, 2008, 2009) and syllogistic reasoning (Morsanyi & Handley, 2012).

Information is also more likely to be accepted as true when it comes from a credible and trustworthy source. As decades of persuasion research illustrates, evaluations of source credibility can be based on declarative information that bears, for example, on the communicator’s expertise, education, achievement, or institutional affiliation and the presence or absence of conflicting interests (for reviews, see Eagly & Chaiken, 1993; Petty & Cacioppo, 1986). However, credibility judgments can also be based on feelings of familiarity. In daily life, people trust familiar others more than strangers (Luhmann, 1979), from personal interactions to e-commerce (Gefen, 2000). Familiarity resulting from previous encounters or even just repeatedly seeing pictures of a face is sufficient to increase perceptions of honesty and sincerity as well as agreement with what the person says (Brown, Brown, & Zoccoli, 2002; Weisbuch & Mackie, 2009). Similarly, the mere repetition of a name can make an unknown name seem familiar, making its bearer “famous overnight” (Jacoby, Woloshyn, & Kelley, 1989), which may also increase perceived expertise. Familiar people are also easier to recognize and remember, and their names become easier to pronounce with repeated encounters. Variables that influence the ease with which source information can be processed can therefore enhance the perceived credibility of the source. Indeed, a given claim is more likely to be judged true when the name of its source is easy to pronounce (Newman et al., 2014).

To assess the likely truth of a claim, people also consider whether others believe it – if many people agree, there’s probably something to it. This social consensus (Festinger, 1954) criterion is central to many social influence processes and is sometimes referred to as the principle of “social proof” (Cialdini, 2009). As numerous studies indicated, people are more confident in their beliefs if they are shared by others (Newcomb, 1943; Visser & Mirabile, 2004), more likely to endorse a message if many others have done so as well (Cialdini, 2009), and place more trust in what they remember if others remember it similarly (Harris & Hahn, 2009; Ross, Buehler, & Karr, 1998). Conversely, perceiving dissent reliably undermines message acceptance, which makes reports on real or fabricated controversies an efficient strategy for swaying public opinion (Lewandowsky, Ecker, Seifert, Schwarz, & Cook, 2012; Lewandowsky, Gignac, & Vaughan, 2013). To assess the extent of consensus, people may consult public opinion polls or ask their friends. Alternatively, they may rely on how “familiar” the belief feels – after all, one should have encountered popular beliefs, shared by many, more frequently than unpopular beliefs, held by few. Empirically, familiar information is easier to read, understand, and remember than unfamiliar information, which makes ease of processing a (fallible) indicator of familiarity and popularity. Accordingly, incidental changes in ease of processing can influence perceived consensus.

Finally, people’s confidence in a belief increases with the amount of supporting evidence. Support can be assessed through an external search, as in a scientific literature review or through recall of pertinent information from memory; in either case, confidence increases with the amount of supportive information. Alternatively, support can be gauged from how easy it is to find supportive evidence – the more evidence there is, the easier it should be to find some (in memory or in the literature). This lay theory is at the heart of Tversky and Kahneman’s (1973) availability heuristic. Unfortunately, this heuristic can be misleading. If the only supportive piece of information comes to mind easily because it has been endlessly repeated or is very vivid and memorable, we may erroneously conclude that support is strong. Moreover, attention to what comes to mind and attention to the ease with which it does so will often lead to different conclusions. On the one hand, reliance on the substantive arguments brought to mind results in higher confidence the more arguments one retrieves or generates. On the other hand, reliance on ease of recall results in lower confidence the more arguments one tries to come up with because finding many arguments is difficult, which suggests that there probably aren’t many (Haddock, Rothman, Reber, & Schwarz, 1999; for reviews, see Schwarz, 1998; Schwarz & Vaughn, 2002).

Regardless of which truth criteria people draw on, easily processed information enjoys an advantage over information that is difficult to p...

Table of contents

• Cover
• Half Title
• Series Page
• Title Page
• Copyright Page
• Contents
• List of contributors
• Introduction
• 1 When (fake) news feels true: Intuitions of truth and the acceptance and correction of misinformation
• 2 Connections
• 3 When Do People Believe Conspiracy Theories?
• 4 The dynamics of groups online
• 5 Belonging in an age of technology
• 6 How do online social networks influence people’s emotional lives?
• 7 Online relationship formation
• 8 Identity citizenship: Authenticity, intersectionality and a new populism