Philosophy of Psychology. Lisa Bortolotti
sometimes implying that if a behavior is adaptive it is rational. […] I think a conflation of these interests is at the heart of the disputes between researchers working in the heuristics and biases tradition and their critics in the evolutionary psychology camp. (Stanovich 2009, 55–56)
Certainly, it could be biologically adaptive to have a cognitive system that is especially good at dealing with frequencies but not very good at dealing with probabilities. But is such a system not only adaptive but also rational? This issue is closely related to the aim or purpose of cognition, which we will discuss in the next section.
Another problem is that the ecological picture seems to imply that humans used to be rational, but they no longer are. The concern, expressed again by Stanovich, is that human reasoning styles probably used to be successful in the ancient environment, but are no longer successful in the current environment, where probabilities tend not to be represented in a frequentist format, and information about cheaters no longer has special biological significance:
Unfortunately, the modern world tends to create situations where some of the default values of evolutionarily adapted cognitive systems are not optimal. […] [M]odern technological societies continually spawn situations where humans must decontextualize information – where they must deal abstractly and in a depersonalized manner with information. (Stanovich 2004, 122)
It is therefore a challenge to defend optimism even when adopting the ecological picture of human rationality. Our temporary conclusion is that none of the objections reviewed so far constitute a serious threat to the pessimistic interpretation of the experimental results. In the next section, we will explore another issue, which is closely related to the dispute between the standard picture and the ecological picture.
1.6 The Aim of Cognition
Aiming at Truth
Why do we accept the standard picture of rationality rather than, say, the ecological picture? The idea that good reasoning aligns with logical and mathematical rules is often associated with the idea that the aim or purpose of cognition is truth; because the aim of cognition is truth, and because logical and mathematical rules are the best guides to truth, human agents ought to reason in accordance with logical and mathematical rules.
But is the purpose of cognition really truth? Does cognition aim at truth? Perhaps Gigerenzer’s fundamental disagreement with the standard picture lies in his dissatisfaction with the idea of truth as the aim of cognition. He does not believe that truth is the aim of cognition, or at least he does not believe that it is the only aim: ‘[A]daptive behavior has other goals than logical truth or consistency, such as dealing intelligently with other people’ (Gigerenzer 2008, 12). From this perspective, human reasoning should not be evaluated in terms of logical and mathematical rules. Rather, it should be evaluated in terms of its contribution to biological and ecological success.
This takes us to the topic of the aim of cognition, which is the last topic in this chapter. We cannot fully address the issue of the aim of cognition here, but we will identify some considerations that need to be taken into account in a full investigation of this issue. As the aim of cognition is too broad a target for our investigation, let us focus on the mental state of believing and explore what its aim might be.
What does the ‘aim’ of belief mean? There could be more than one correct answer to this question, but our tentative answer is that the ‘aim’ of belief is the function of the belief-forming mechanisms (i.e., the mechanisms that produce beliefs in response to some inputs, including perceptual inputs and other beliefs). To say that the aim of belief is truth is to say that the belief-forming mechanisms have the function of tracking truth (i.e., producing true beliefs, avoiding false ones, etc.) (e.g., Velleman 2000; see also Miyazono 2018).
By ‘function’ we mean so-called ‘aetiological’ function; on this understanding, function aligns with evolutionary design (e.g., Millikan 1989; Neander 1991). In other words, X has the aetiological function of doing Y if Xs were selected for doing Y in their evolutionary history. For example, hearts have the aetiological function of pumping blood: hearts were selected for pumping blood in their evolutionary history. Again, kidneys have the aetiological function of filtering metabolic wastes from blood: kidneys were selected for filtering metabolic wastes from blood in their evolutionary history. Similarly, belief-forming mechanisms have the aetiological function of tracking truth if the mechanisms were selected for tracking truth in their evolutionary history.
Thus, our questions are: ‘Do belief-forming mechanisms really have the function of tracking truth?’ and ‘Were they selected for tracking truth in their evolutionary history?’
There are two tempting arguments against the idea that belief-forming mechanisms have the function of tracking truth. Both arguments, in our view, are problematic.
The first argument is that the function of belief-forming mechanisms cannot be tracking truth because ‘natural selection does not care about truth; it cares only about reproductive success’ (Stich 1990, 62).
This argument might seem compelling at first glance, but it does not stand careful scrutiny. The problem with this argument becomes obvious when we run a parallel argument about hearts. The function of hearts cannot be pumping blood because natural selection does not care about pumping blood; natural selection only cares about reproductive success. The weakness of the parallel argument is clear: the choice between pumping blood and reproductive success as the function of hearts is a false dichotomy. After all, hearts contributed to reproductive success by pumping blood, which is why the function of hearts is to pump blood. Similarly, the choice between tracking truth and reproductive success as the function of belief-forming mechanisms is a false dichotomy. Perhaps belief-forming mechanisms contributed to reproductive success by tracking truth, in which case the function of the mechanisms is to track truth (Millikan 2009; Wilkins & Griffiths 2012).
The second argument is as follows: the function of belief-forming mechanisms cannot be tracking truth because, as we have seen in the heuristics and biases studies, such mechanisms – and reasoning mechanisms especially – lead to systematic errors and biases. In other words, systematic errors and biases constitute evidence that tracking truth is not the function of belief-forming mechanisms.
This argument is not compelling either. Systematically failing to do X is perfectly consistent with having the function of doing X. For example, sperm has the function of fertilizing an egg, although most sperm fail to fulfil their function. Although most sperm fail to fertilize an egg, some sperm do fertilize an egg, and that is how sperm has contributed to reproductive success in its evolutionary history. Similarly, it is possible that belief-forming mechanisms have the function of tracking truth, despite them systematically producing false beliefs. Although these mechanisms often produce false beliefs, it is possible that they do produce true beliefs in many other cases, and that is how belief-forming mechanisms have contributed to survival and reproductive success, at least in their evolutionary history. Let us put this a different way. Most sperm do not fertilize an egg, not because sperm do not aim at fertilizing an egg but because some failure can be tolerated as long as some sperm successfully fertilize an egg in biologically crucial contexts. Similarly, human belief-forming mechanisms exhibit systematic errors and biases, not because belief-forming mechanisms do not aim at tracking truth but because some errors and biases (even systematic ones) can be tolerated as long as some beliefs successfully track truth in biologically crucial contexts.
To sum up, the arguments above fail to refute the idea that belief-forming mechanisms have the function of tracking truth. In other words, the idea that truth is the aim of belief has not been ruled out yet. Before closing this chapter, we would like to mention a particularly relevant discussion by Ryan McKay and Daniel Dennett (2009).
Positive Illusions
McKay and Dennett endorse the view that in most cases adaptive beliefs are true beliefs. But there is an exception: positive illusions (i.e., optimistically biased judgments about oneself [see Box