Do children think that duplicating the body also duplicates the mind?

Abstract

Philosophers use hypothetical duplication scenarios to explore intuitions about personal identity. Here we examined 5- to 6-year-olds’ intuitions about the physical properties and memories of a live hamster that is apparently duplicated by a machine. In Study 1, children thought that more of the original’s physical properties than episodic memories were present in the duplicate hamster. In Study 2, children thought that episodic memories of the hamster were less likely to duplicate than events captured by a digital camera. Studies 3 and 4 ruled out lower-level explanations of these effects. Study 5 showed that naming the original hamster further reduced the inferred duplication of memories in the second hamster. Taken together, these studies are consistent with the view that young children think that some mental properties are distinct from physical ones.

Introduction

Philosophers have long been intrigued by perfect duplicates. In The Transfiguration of the Commonplace, Arthur Danto wrote, “I have been obsessed with paired cases where only one member of the couple is an artwork.” (1981, p. 90). An example of this is Marcel Duchamp’s Fountain, an important and much-admired artwork, which Danto imagined paired with a perceptually indistinguishable urinal, an everyday object that has no artistic merit. Or one might compare an original Picasso with a perfect forgery – although the perceptual and microphysical properties of the two objects might be identical, still, they differ greatly in value and (on some analyses) aesthetic merit. Such examples illustrate the limits of physical properties in the categorisation and evaluation of art, and suggest further that beliefs about the works’ history have a particular relevance (see also Bloom, 2010, Hood, 2009).

Consider now the duplication of people – a classic philosopher’s thought experiment used to contrast and develop theories of personal identity and psychological states (Armstrong, 1980, Nozick, 1981, Shoemaker, 1979; see Martin, Barresi, and Giovanelli (1998) for historical review). In one well-known scenario developed by Derek Parfit, Parfit enters a futuristic teletransporter pod that scans his body to record the exact state of all physical matter and then sends this information to Mars, where a duplicate is re-constructed from this information with new matter (Parfit, 1984). Back on Earth, the original Parfit is destroyed. Is the person who now exists on Mars really Parfit? Does he (or it) share the same mental states as the original?

The focus of the experiments reported below is not the metaphysical issues, but rather on our developing folk psychological intuitions. We ask whether, in a Parfit-like scenario, young children will believe that certain mental states will be shared between an original and a duplicate. Will they believe that their bodies will be identical but their minds will not be?

Previous research suggests that, like adults, children appreciate that one can be the same individual despite changes in personal properties – for instance, one can change clothing, cut his hair, and lose some weight and yet still remain the same person (see e.g., Gutheil & Rosengren, 1996). Children know as well that a proper name refers to an individual that has a unique spatiotemporal path and does not extend to an individual of identical appearance in a different location (MacNamara, 1982, Sorrentino, 2001; see also Gutheil, Gelman, Klein, Michos, & Kelaita, 2008). But children have not yet been tested on Parfit-like instances in which an apparent duplicate is not connected to the original through a distinct spatio-temporal path (though see Rips, Blok, and Newman (2006) and Rhemtulla and Hall (2009) for research with adults).

One motivation to ask this question is to explore the claim that people are natural Cartesian dualists. It is sometimes argued that, prior to exposure to scientific education, children assume that mental properties are distinct from physical ones (Bering and Bjorklund, 2004, Bloom, 2004, Cohen, 2007, Johnson and Wellman, 1982). The duplication scenario can explore this. If psychological states emerge from (supervene upon) physical brain states, then the duplicate should have all the same mental states as the original. However, if one is a Cartesian dualist, believing that the experiencing self is distinct from the body, then a physically identical duplicate might have different mental states.

Rather than a strictly verbal thought experiment, we tested children by showing them actual duplication. In previous research, we found that young children accept the possibility of duplication when presented with a scientific looking machine that can apparently copy real objects, such as toys (Hood & Bloom, 2008). We use this method here to duplicate live hamsters. Unlike the Parfit scenario, we did not destroy (or pretend to destroy) the original hamster, in part because of concerns about children’s sensibilities and in part because the extent to which children attribute physical and mental properties to the original serves as a baseline with which to compare their attribution to the duplicate.

We focus on 5- and 6-year-olds because we found in pilot studies that younger children often perseverated on a single response and failed control questions while older children were less credulous of the duplicating machine. Further, this age group overlaps with studies that use alternative methods to explore young children’s understanding of bodies and minds (e.g., Choe, Keil, & Bloom, 2012, Gottfried et al., 1999, Johnson, 1990).

We restricted the investigation of mental states to knowledge states based on episodic memories. From a methodological perspective, this allowed us to experimentally manipulate the presence or absence of mental states in a way that was easily understood by children; it also enabled us to compare mental states of an animal with the internal states of a camera.

We also focused on episodic memory because it may have a particular relevance to personal identity. According to Locke’s (1690/1975, chap. xxvii) theory of personal identity, individuality is defined in terms of the continuity of memories. Personal identity may also involve other mental states such as preferences, attitudes, and emotions, but to a lesser extent. For example, two very different individuals can both like vanilla ice-cream or believe in ghosts and a single individual can change her mind about ice-cream and ghosts without changing her identity. In contrast, experiencing an event or being supplied with exclusive knowledge induces novel epistemic mental states that are not typically shared with others who were not present at the induction. Note also that our focus on epistemic states is similar to the work of Gutheil and colleagues (Gutheil et al., 2008), who asked children whether two identical looking Winnie-the-Pooh dolls shared the same knowledge state that was supplied to only one doll. Similarly, in the experiments reported here, we introduced specific events that we could ensure were understood by the children to constitute memories for the original hamster.