Hoerl & McCormack (H&M) propose that humans develop two systems for representing the world: one that possesses a temporal dimension (and therefore can be used to reason about time) and one that lacks a temporal dimension (and therefore cannot be used to reason about time). Although H&M's proposal is one of many to posit two systems of human cognition (i.e. dual-systems theories) (Evans Reference Evans2008; Melnikoff & Bargh Reference Melnikoff and Bargh2018), it is nonetheless highly original. To our knowledge, it is the first dual-systems theory to define its systems in terms of possessing versus lacking a single representational dimension. Most other dual-systems theories define their systems in terms of neural substrates (e.g., neocortex vs. hippocampus), modes of processing (e.g., associative vs. propositional), and/or operating characteristics (e.g., automatic vs. controlled). Although we appreciate H&M's innovation, we are skeptical about their dual-systems approach to temporal cognition. A preferable approach, we will argue, would be to posit that the ability to reason about time emerges from the temporal enrichment of a single system (Keren & Schul Reference Keren and Schul2009; Kruglanski & Gigerenzer Reference Kruglanski and Gigerenzer2011; Osman Reference Osman2004).
H&M's evidence for dual systems is that people simultaneously affirm two contradictory beliefs about time: (i) only the present exists, and (ii) the past, present, and future all exist equally. H&M assume that a single system cannot affirm contradictory beliefs simultaneously, and therefore conclude that people's beliefs about time are affirmed by multiple systems. This type of argument was pioneered by Sloman (Reference Sloman1996), whom H&M follow by never explaining precisely why a single system cannot simultaneously affirm contradictory beliefs. What is the problem supposed to be? Presumably, a single system can operate according to the rules: “If condition x is satisfied then affirm the belief p” and “If condition y is satisfied then affirm the belief not-p,” where x and y are not mutually exclusive. If x and y were satisfied simultaneously, would such a system not simultaneously affirm p and not-p? We submit that it would, and hence, simultaneous contradictory beliefs do not recommend dual systems (Bermúdez Reference Bermúdez2000; Betsch & Fiedler Reference Betsch and Fiedler1999; Ferguson et al. Reference Ferguson, Mann, Wojnowicz, Sherman, Gawronski and Trope2014; Gigerenzer & Regier Reference Gigerenzer and Regier1996).
Let us be more concrete. Suppose that a system determines whether things exist using two rules: the present rule and the causality rule. The present rule states that only temporally present things exist. The causality rule states that things exist if and only if they can affect or be affected by temporally present things. Both rules seem plausible enough. A common-sense list of things that exist might include Oprah Winfrey, Stonehenge, and salami (all present things that can affect or be affected by present things) but would not include Sherlock Holmes, Shakespeare's original Globe Theatre, or Martian outposts (all non-present things that cannot affect or be affected by present things). Notice, however, that the two rules disagree about the existence of the past, present, and future. The present rule entails that the present is all that exists. Yet the causality rule entails that the past, present, and future all exist equally (because the past affects the present and the present affects the future). Therefore, the simultaneous application of the present rule and the causality rule by a single system would produce the simultaneous contradictory beliefs that motivated H&M's dual-process approach in the first place.
One might question our line of reasoning by arguing that the human mind cannot affirm the conjunction of two contradictory beliefs (i.e., a single belief of the form p & not-p). Indeed, it is hard to imagine someone genuinely affirming the conjunctive belief that only the present exists and the past, present, and future all exist equally. But nothing like this needs to occur when a single system simultaneously affirms contradictory beliefs. The simultaneous affirmation of contradictory beliefs does not entail the affirmation of the conjunction of those beliefs (Bermúdez Reference Bermúdez2000). The statements “S believes p at time t” and “S believes not-p at time t” do not jointly entail that the S ever has a single belief of the form p & not-p. Accordingly, a human mind that is incapable of affirming conjunctions of contradictory beliefs can still (within a single system) affirm contradictory beliefs simultaneously. We doubt, therefore, that the simultaneous holding of contradictory beliefs requires dual systems.
Having refuted H&M's evidence for their dual-systems theory, we hold in favor of the alternative hypothesis that temporal reasoning abilities emerge from the temporal enrichment of a single system. We suggest that this alternative hypothesis requires fewer assumptions and auxiliary mechanisms than H&M's dual-systems approach, and therefore is (necessarily) less likely to be wrong. The problem is that an agent with both a temporal reasoning system (i.e. a temporally enriched mental model) and a temporal updating system (i.e., a temporally impoverished mental model) has two systems with which to reason about all non-temporal dimensions. Such an agent could use either system to reason about distance, for example, because distance is represented in both systems. This raises a number of questions. Which system would such an agent use to compute its shortest route home, and what would determine this? If the agent used one system to compute its shortest route home, would the other system have access to the answer? If so, how would information transfer between the two systems? To be able to answer these questions H&M must specify how the systems are selected for processing non-temporal information, and how non-temporal information transfers between the two systems, if at all. Doing so will require H&M to make additional assumptions and to posit additional mechanisms, all of which have the possibility of being false, necessarily rendering the dual-systems theory less plausible than the alternative hypothesis that temporal reasoning abilities emerge from the temporal enrichment of a single system.
You have
Access
Hoerl & McCormack (H&M) propose that humans develop two systems for representing the world: one that possesses a temporal dimension (and therefore can be used to reason about time) and one that lacks a temporal dimension (and therefore cannot be used to reason about time). Although H&M's proposal is one of many to posit two systems of human cognition (i.e. dual-systems theories) (Evans Reference Evans2008; Melnikoff & Bargh Reference Melnikoff and Bargh2018), it is nonetheless highly original. To our knowledge, it is the first dual-systems theory to define its systems in terms of possessing versus lacking a single representational dimension. Most other dual-systems theories define their systems in terms of neural substrates (e.g., neocortex vs. hippocampus), modes of processing (e.g., associative vs. propositional), and/or operating characteristics (e.g., automatic vs. controlled). Although we appreciate H&M's innovation, we are skeptical about their dual-systems approach to temporal cognition. A preferable approach, we will argue, would be to posit that the ability to reason about time emerges from the temporal enrichment of a single system (Keren & Schul Reference Keren and Schul2009; Kruglanski & Gigerenzer Reference Kruglanski and Gigerenzer2011; Osman Reference Osman2004).
H&M's evidence for dual systems is that people simultaneously affirm two contradictory beliefs about time: (i) only the present exists, and (ii) the past, present, and future all exist equally. H&M assume that a single system cannot affirm contradictory beliefs simultaneously, and therefore conclude that people's beliefs about time are affirmed by multiple systems. This type of argument was pioneered by Sloman (Reference Sloman1996), whom H&M follow by never explaining precisely why a single system cannot simultaneously affirm contradictory beliefs. What is the problem supposed to be? Presumably, a single system can operate according to the rules: “If condition x is satisfied then affirm the belief p” and “If condition y is satisfied then affirm the belief not-p,” where x and y are not mutually exclusive. If x and y were satisfied simultaneously, would such a system not simultaneously affirm p and not-p? We submit that it would, and hence, simultaneous contradictory beliefs do not recommend dual systems (Bermúdez Reference Bermúdez2000; Betsch & Fiedler Reference Betsch and Fiedler1999; Ferguson et al. Reference Ferguson, Mann, Wojnowicz, Sherman, Gawronski and Trope2014; Gigerenzer & Regier Reference Gigerenzer and Regier1996).
Let us be more concrete. Suppose that a system determines whether things exist using two rules: the present rule and the causality rule. The present rule states that only temporally present things exist. The causality rule states that things exist if and only if they can affect or be affected by temporally present things. Both rules seem plausible enough. A common-sense list of things that exist might include Oprah Winfrey, Stonehenge, and salami (all present things that can affect or be affected by present things) but would not include Sherlock Holmes, Shakespeare's original Globe Theatre, or Martian outposts (all non-present things that cannot affect or be affected by present things). Notice, however, that the two rules disagree about the existence of the past, present, and future. The present rule entails that the present is all that exists. Yet the causality rule entails that the past, present, and future all exist equally (because the past affects the present and the present affects the future). Therefore, the simultaneous application of the present rule and the causality rule by a single system would produce the simultaneous contradictory beliefs that motivated H&M's dual-process approach in the first place.
One might question our line of reasoning by arguing that the human mind cannot affirm the conjunction of two contradictory beliefs (i.e., a single belief of the form p & not-p). Indeed, it is hard to imagine someone genuinely affirming the conjunctive belief that only the present exists and the past, present, and future all exist equally. But nothing like this needs to occur when a single system simultaneously affirms contradictory beliefs. The simultaneous affirmation of contradictory beliefs does not entail the affirmation of the conjunction of those beliefs (Bermúdez Reference Bermúdez2000). The statements “S believes p at time t” and “S believes not-p at time t” do not jointly entail that the S ever has a single belief of the form p & not-p. Accordingly, a human mind that is incapable of affirming conjunctions of contradictory beliefs can still (within a single system) affirm contradictory beliefs simultaneously. We doubt, therefore, that the simultaneous holding of contradictory beliefs requires dual systems.
Having refuted H&M's evidence for their dual-systems theory, we hold in favor of the alternative hypothesis that temporal reasoning abilities emerge from the temporal enrichment of a single system. We suggest that this alternative hypothesis requires fewer assumptions and auxiliary mechanisms than H&M's dual-systems approach, and therefore is (necessarily) less likely to be wrong. The problem is that an agent with both a temporal reasoning system (i.e. a temporally enriched mental model) and a temporal updating system (i.e., a temporally impoverished mental model) has two systems with which to reason about all non-temporal dimensions. Such an agent could use either system to reason about distance, for example, because distance is represented in both systems. This raises a number of questions. Which system would such an agent use to compute its shortest route home, and what would determine this? If the agent used one system to compute its shortest route home, would the other system have access to the answer? If so, how would information transfer between the two systems? To be able to answer these questions H&M must specify how the systems are selected for processing non-temporal information, and how non-temporal information transfers between the two systems, if at all. Doing so will require H&M to make additional assumptions and to posit additional mechanisms, all of which have the possibility of being false, necessarily rendering the dual-systems theory less plausible than the alternative hypothesis that temporal reasoning abilities emerge from the temporal enrichment of a single system.