Gilead et al. describe a new, exciting theory of abstraction, which is a step forward for work on this interdisciplinary topic. Among other contributions, their theory will allow researchers across fields to use a common terminology and better appreciate insights from other fields.
Two key elements that need to be integrated into their account are (1) an appreciation of how feelings such as fluency influence mental simulation, and (2) an appreciation of how prediction errors persist even among people with substantial personal experience. Both of these elements could be integrated into Gilead and colleagues' account, though it would be worthwhile delineating how.
Gilead et al. argue that people construe objects abstractly when these objects are psychologically distant and when it is rational to do so based on detail/accuracy trade-offs. Yet, feelings such as fluency also influence construal level and mental simulation. People are more likely to simulate events when they come to mind easily (Kahneman & Tversky Reference Kahneman, Tversky, Kahneman, Slovic and Tversky1982; Schwarz et al. Reference Schwarz, Bless, Strack, Klumpp, Rittenauer-Schatka and Simons1991).
Feelings such as fluency and mood influence mental construal. In everyday life, people attend to their feelings as a source of information, and different feelings provide different types of information and trigger different processing strategies (Schwarz & Clore Reference Schwarz and Clore1996). For example, being in a negative (vs. positive) mood can spontaneously generate a detail-oriented elaboration, which is usually adaptive in problematic situations (Schwarz Reference Schwarz, Martin and Clore2001). A negative mood, by signaling not only a problem but also its imminence, can lead people to adopt a concrete construal and focus attention on the immediate, which can increase how much immediate events are prioritized (Labroo & Patrick Reference Labroo and Patrick2009; Mrkva & Van Boven Reference Mrkva and Van Boven2017). In contrast, positive mood, by signaling that the immediate environment is benign, encourages an abstract construal and a long-term perspective. Further findings (see, e.g., Bless et al. Reference Bless, Clore, Schwarz, Golisano, Rabe and Wölk1996; Gardner et al. Reference Gardner, Wansink, Kim and Park2014) corroborate the idea that a negative mood elicits proximal, concrete construals whereas a positive mood often elicits distal, abstract construals. In addition to positive or negative mood, both emotional arousal and fluency reduce feelings of psychological distance and influence mental simulations (Alter & Oppenheimer Reference Alter and Oppenheimer2009; Mrkva et al. Reference Mrkva, Travers and Van Boven2018; Szpunar & Schacter Reference Szpunar and Schacter2013; Van Boven et al. Reference Van Boven, Kane, McGraw and Dale2010). Bodily sensations, such as upward or downward head and eye movements can also influence simulations and mental construals, as when looking upward evokes more abstract imagery of distant objects (Barsalou Reference Barsalou1999; Cian Reference Cian2017; Van Kerckhove et al. Reference Van Kerckhove, Geuens and Vermeir2014).
Because feelings such as fluency influence mental simulation, people's feelings can make them more likely to simulate events even when fluent events are objectively less likely to occur. For example, people might simulate an extremely unlikely terrorist attack when it is top-of-mind and fluent, whereas failing to simulate more likely risks for which they have equivalent information and experience (Mrkva et al. Reference Mrkva, Travers and Van Boven2018; Sherman et al. Reference Sherman, Cialdini, Schwartzman and Reynolds1985; Sunstein Reference Sunstein2003). Doing so can have drastic consequences, as when Americans decided to drive rather than fly shortly after 9/11, likely resulting in hundreds of additional deaths from traffic accidents (Blalock et al. Reference Blalock, Kadiyali and Simon2009).
Finally, feelings can lead to empathy gaps and other systematic prediction errors. Empathy gaps prevent people from effectively simulating other mental states, even when they possess abundant relevant conceptual knowledge. For example, when in neutral or satiated states, students do not appreciate how much hunger would influence preferences (Van Boven & Loewenstein Reference Van Boven and Loewenstein2003), smokers fail to appreciate how strong their desires to smoke will be (Sayette et al. Reference Sayette, Loewenstein, Griffin and Black2008), and drug addicts underestimate how alluring their urges will be (Van Boven et al. Reference Van Boven, Loewenstein, Dunning and Nordgren2013). These empathy gaps persist even amid substantial personal experience, considering that smokers and drug users typically have vast experience with the relevant concreta and abstracta (Van Boven et al. Reference Van Boven, Loewenstein, Dunning and Nordgren2013). They also occur even when people in neutral “cold” states and emotional “hot” states have the same objective contextual details about an event (Van Boven et al. Reference Van Boven, Loewenstein, Dunning and Nordgren2013). In particular, people may have difficulty translating experiences in an emotional, “hot” state to an abstract lesson implemented in a “cold” state (and vice versa; Loewenstein Reference Loewenstein1996; Loewenstein et al. Reference Loewenstein, O'Donoghue and Rabin2003; Van Boven & Loewenstein Reference Van Boven and Loewenstein2003).
Another systematic prediction error, the planning fallacy, persists in spite of substantial personal experience (Buehler et al. Reference Buehler, Griffin and Ross1994; Reference Buehler, Griffin, Ross, Gilovich, Griffin and Kahneman2002). People underestimate how long tasks will take to complete and how much money they will spend, even if they are experienced with these decisions and have made many similar underestimation errors in the past (Buehler et al. Reference Buehler, Griffin, Ross, Gilovich, Griffin and Kahneman2002; Griffin & Buehler Reference Griffin and Buehler2005; Peetz & Buehler Reference Peetz and Buehler2009). These errors partly reflect tendencies to rely on hope over experience, simulate the easiest-to-imagine, best-case scenario, and ignore other tasks and obstacles that may interfere (Buehler et al. Reference Buehler, Griffin, Ross, Gilovich, Griffin and Kahneman2002; Buehler & Griffin Reference Buehler and Griffin2003; Kahneman & Lovallo Reference Kahneman and Lovallo1993). Gilead et al. state that people should (and do) simulate events when they have past experience in their reservoir of memories. Yet, it is important to note that people are prone to systematic prediction errors such as empathy gaps and the planning fallacy despite having previous experience. The observation that empathy gaps and other prediction-reality gaps persist even among people with vast experience offers challenges for any predictive cognition framework that focuses exclusively on experience and conceptual knowledge.
In sum, Gilead et al. develop a rich framework, but need to allow ample room in their framework for feelings (emotions, fluency, and other states). These feelings can shape whether people use abstract thought, how they are impacted by simulations, and whether their predictions will correspond to reality.
Gilead et al. describe a new, exciting theory of abstraction, which is a step forward for work on this interdisciplinary topic. Among other contributions, their theory will allow researchers across fields to use a common terminology and better appreciate insights from other fields.
Two key elements that need to be integrated into their account are (1) an appreciation of how feelings such as fluency influence mental simulation, and (2) an appreciation of how prediction errors persist even among people with substantial personal experience. Both of these elements could be integrated into Gilead and colleagues' account, though it would be worthwhile delineating how.
Gilead et al. argue that people construe objects abstractly when these objects are psychologically distant and when it is rational to do so based on detail/accuracy trade-offs. Yet, feelings such as fluency also influence construal level and mental simulation. People are more likely to simulate events when they come to mind easily (Kahneman & Tversky Reference Kahneman, Tversky, Kahneman, Slovic and Tversky1982; Schwarz et al. Reference Schwarz, Bless, Strack, Klumpp, Rittenauer-Schatka and Simons1991).
Feelings such as fluency and mood influence mental construal. In everyday life, people attend to their feelings as a source of information, and different feelings provide different types of information and trigger different processing strategies (Schwarz & Clore Reference Schwarz and Clore1996). For example, being in a negative (vs. positive) mood can spontaneously generate a detail-oriented elaboration, which is usually adaptive in problematic situations (Schwarz Reference Schwarz, Martin and Clore2001). A negative mood, by signaling not only a problem but also its imminence, can lead people to adopt a concrete construal and focus attention on the immediate, which can increase how much immediate events are prioritized (Labroo & Patrick Reference Labroo and Patrick2009; Mrkva & Van Boven Reference Mrkva and Van Boven2017). In contrast, positive mood, by signaling that the immediate environment is benign, encourages an abstract construal and a long-term perspective. Further findings (see, e.g., Bless et al. Reference Bless, Clore, Schwarz, Golisano, Rabe and Wölk1996; Gardner et al. Reference Gardner, Wansink, Kim and Park2014) corroborate the idea that a negative mood elicits proximal, concrete construals whereas a positive mood often elicits distal, abstract construals. In addition to positive or negative mood, both emotional arousal and fluency reduce feelings of psychological distance and influence mental simulations (Alter & Oppenheimer Reference Alter and Oppenheimer2009; Mrkva et al. Reference Mrkva, Travers and Van Boven2018; Szpunar & Schacter Reference Szpunar and Schacter2013; Van Boven et al. Reference Van Boven, Kane, McGraw and Dale2010). Bodily sensations, such as upward or downward head and eye movements can also influence simulations and mental construals, as when looking upward evokes more abstract imagery of distant objects (Barsalou Reference Barsalou1999; Cian Reference Cian2017; Van Kerckhove et al. Reference Van Kerckhove, Geuens and Vermeir2014).
Because feelings such as fluency influence mental simulation, people's feelings can make them more likely to simulate events even when fluent events are objectively less likely to occur. For example, people might simulate an extremely unlikely terrorist attack when it is top-of-mind and fluent, whereas failing to simulate more likely risks for which they have equivalent information and experience (Mrkva et al. Reference Mrkva, Travers and Van Boven2018; Sherman et al. Reference Sherman, Cialdini, Schwartzman and Reynolds1985; Sunstein Reference Sunstein2003). Doing so can have drastic consequences, as when Americans decided to drive rather than fly shortly after 9/11, likely resulting in hundreds of additional deaths from traffic accidents (Blalock et al. Reference Blalock, Kadiyali and Simon2009).
Finally, feelings can lead to empathy gaps and other systematic prediction errors. Empathy gaps prevent people from effectively simulating other mental states, even when they possess abundant relevant conceptual knowledge. For example, when in neutral or satiated states, students do not appreciate how much hunger would influence preferences (Van Boven & Loewenstein Reference Van Boven and Loewenstein2003), smokers fail to appreciate how strong their desires to smoke will be (Sayette et al. Reference Sayette, Loewenstein, Griffin and Black2008), and drug addicts underestimate how alluring their urges will be (Van Boven et al. Reference Van Boven, Loewenstein, Dunning and Nordgren2013). These empathy gaps persist even amid substantial personal experience, considering that smokers and drug users typically have vast experience with the relevant concreta and abstracta (Van Boven et al. Reference Van Boven, Loewenstein, Dunning and Nordgren2013). They also occur even when people in neutral “cold” states and emotional “hot” states have the same objective contextual details about an event (Van Boven et al. Reference Van Boven, Loewenstein, Dunning and Nordgren2013). In particular, people may have difficulty translating experiences in an emotional, “hot” state to an abstract lesson implemented in a “cold” state (and vice versa; Loewenstein Reference Loewenstein1996; Loewenstein et al. Reference Loewenstein, O'Donoghue and Rabin2003; Van Boven & Loewenstein Reference Van Boven and Loewenstein2003).
Another systematic prediction error, the planning fallacy, persists in spite of substantial personal experience (Buehler et al. Reference Buehler, Griffin and Ross1994; Reference Buehler, Griffin, Ross, Gilovich, Griffin and Kahneman2002). People underestimate how long tasks will take to complete and how much money they will spend, even if they are experienced with these decisions and have made many similar underestimation errors in the past (Buehler et al. Reference Buehler, Griffin, Ross, Gilovich, Griffin and Kahneman2002; Griffin & Buehler Reference Griffin and Buehler2005; Peetz & Buehler Reference Peetz and Buehler2009). These errors partly reflect tendencies to rely on hope over experience, simulate the easiest-to-imagine, best-case scenario, and ignore other tasks and obstacles that may interfere (Buehler et al. Reference Buehler, Griffin, Ross, Gilovich, Griffin and Kahneman2002; Buehler & Griffin Reference Buehler and Griffin2003; Kahneman & Lovallo Reference Kahneman and Lovallo1993). Gilead et al. state that people should (and do) simulate events when they have past experience in their reservoir of memories. Yet, it is important to note that people are prone to systematic prediction errors such as empathy gaps and the planning fallacy despite having previous experience. The observation that empathy gaps and other prediction-reality gaps persist even among people with vast experience offers challenges for any predictive cognition framework that focuses exclusively on experience and conceptual knowledge.
In sum, Gilead et al. develop a rich framework, but need to allow ample room in their framework for feelings (emotions, fluency, and other states). These feelings can shape whether people use abstract thought, how they are impacted by simulations, and whether their predictions will correspond to reality.