4 Reading minds/controlling minds Much of the interest and anxiety provoked by new and developing neuroscientific technologies is centered around two issues: the extent to which these technologies might allow their users to read the thoughts of people, and (as if that prospect was not disturbing enough) the extent to which these technologies might actually be used to control people. Some commentators believe that one or both of these issues are pressing, in the sense that the relevant technolo- gies will soon be available; some even believe that these technologies already exist. In this chapter, we will ask how worried we should be. Are these technologies imminent? And if they are, are they as threatening as they appear? mind reading and mind controlling There has been a great deal of interest in the possibility of ‘‘brain reading’’ as a lie detection technology. The problems with existing lie detectors are well known: they produce high rates both of false positives and of false negatives, and they can be ‘‘beaten’’ by people who deliberately heighten their responses to control questions, which are used to establish a baseline for comparison. In its overview of current lie-detection techniques, the US National Research Council concluded that there is ‘‘little basis for the expectation that a polygraph test could have extremely high accuracy’’ (National Research Council 2003: 212). The reasons for this conclusion are many: because the responses measured are not uniquely involved in deception, because they include responses that can be deliberately controlled and because the technology is difficult to implement in the real-world. The authors conclude that further research and investment can be expected to produce only modest gains in accu- racy. For these kinds of reasons, conventional polygraph results are inadmissible as evidence in most jurisdictions. It is recognition of the severe limitations of current lie detection technologies that is responsible, in important part, for the current interest in lie detectors that directly ‘‘read’’ the mind. Polygraph machines are sensitive to changes in somatic states such as skin conductance (changes in the resistance of the skin to elec- tricity, which is a good indicator of increased sweating), heart rate and blood pressure. Now, the problem with these measures is that, at best, they are only correlated with deliberate deception. They are indications of increased nervousness, and can be manipulated. Moreover, even with a naı ¨ ve subject who does not know how to manipulate these responses, the correlation between the responses and deliberate deception is far from perfect; hence the false positives (where these responses peak but the subject is not lying), and the false negatives (where the subject is lying, but their somatic responses do not reflect this fact). Proponents of neurologically based lie detection argue, apparently plausibly, that such technolo- gies would not be subject to these limitations. It might be possible to fool a system that measures only stand-ins for lies, but it is not possible to fool a system capable of honing in on the lies them- selves. Of course, lies are not directly detectable, but the technology might be capable of doing the next best thing. Thoughts are realized neurologically; in the jargon of cognitive science, they have neural correlates. The correlation between the lie and its neural correlate is, by definition, perfect. Hence, if the lie detection technology can hone in on the neural correlates of lies, it will not give false positives or false negatives, and it will not be able to be fooled. Brains do not lie. The problem with this line of thought is obvious. Though it is certainly true that every thought has a neural correlate – to say that is just to say that thoughts are realized in brains, so the claim should reading minds/controlling minds 134 be uncontroversial to anyone who does not believe in substance dualism – it does not follow that for every type of thought there is a distinct neural correlate. Indeed, the latter claim is rather implau- sible, given that there are endless ways in which thoughts might be categorized (‘‘concerning animals with fewer than four legs,’’ ‘‘con- cerning either my maiden aunt or a ming vase,’’ ‘‘concerning an odd number of elephants’’ – the categories are limited only by our inge- nuity). But the idea that we can detect lies by honing in on their neural correlates requires that to the thought-type ‘‘deliberate deception’’ there corresponds one, or at any rate a tractable number, of distinct neural correlates. Unless and until we develop the tech- nology to translate neural correlates into thoughts (a prospect we shall consider later in this chapter), we can only hope to detect lies if we discover that there is something distinctive about deliberate deception; something we can look for in brain scans. So reliable neurological detection depends upon there being something neurologically distinctive about deliberate deception. Is there? There are various proposals for such distinctive correlates, either of the entire class of deliberate deception, or of various smaller segments of that class. Perhaps the best-known technique of neuro- logical lie detection, ‘‘brain fingerprinting,’’ targets one possible distinctive correlate. Brain fingerprinting does not aim to detect any and all deliberate deception; instead, it is aimed at detecting guilty knowledge. The technology uses memory and encoding related multifaceted electroencephalographic response (MERMER), which combines electroencephalography (EEG) data from several sites on the scalp. Electrodes are attached to the subject’s scalp at several sites, and brain activity is measured while the subject is asked questions or shown pictures. If the subject recognizes a picture or is familiar with the content of the question, they are supposed to exhibit a characteristic response, the P300 wave (so called because it occurs 300 milliseconds after the stimulus). If they do not have the relevant knowledge, the amplitude of the P300 wave will be significantly smaller. mind reading/controlling minds 135 MERMER and brain fingerprinting are the brainchildren of Lawrence Farwell; Farwell has aggressively commercialized the technology, through his company Brain Fingerprinting Laboratories. Farwell claims that brain fingerprinting has several advantages over polygraphy. Because the P300 wave is involuntary, it does not depend upon the cooperation of the person being tested (with one exception: the subject must sit still during testing). This makes it especially suitable for testing people who might be planning a crime, for instance suspected terrorists. The tester might show them bomb- making equipment, and examine their EEG. Most importantly, Far- well claims that brain fingerprinting cannot be manipulated. Whereas clever criminals can beat conventional lie-detection tests, they cannot control the involuntary P300 response. In one test of the technology, subjects were instructed to attempt to conceal the knowledge being probed; nevertheless the guilty knowledge was detected. There were no false positives, false negatives or inde- terminate cases (Farwell and Smith 2001). Farwell and colleagues claim these kinds of results are a spec- tacular vindication of brain fingerprinting. However, there are a number of problems with the technology. First, the method of ana- lysis used by Farwell is proprietary and undisclosed; for that reason there cannot be independent testing of its validity. What few tests there are for brain fingerprinting come exclusively from Farwell’s laboratory. Independent testing is, of course, the gold standard of good science; in its absence, we are entitled to a high degree of scepticism (Wolpe et al. 2005). Moreover, even assuming that Far- well’s tests have been scrupulously conducted, and his own invest- ment in the technique has had no effect in biasing his results (consciously or unconsciously), his sample sizes are too small to yield a great deal of confidence. Moreover, there are difficulties concerning the ecological validity of the technology; that is, the extent to which the findings can be generalized to the world outside the laboratory. It requires carefully controlled situations; in particular, it requires that the reading minds/controlling minds 136 tester possess information that they know will be available only to the perpetrator of the crime. Sometimes such information will be known, but probably only in the minority of cases (Tancredi 2004). Suppose the subject exhibits a P300 to images of a terrorist training camp. Should we conclude that they have trained as a terrorist, or that they watch CNN? In fact, in the only field study to date on the technology, it performed at around chance accuracy (Miyake et al. 1993). Finally, it appears that Farwell may well be wrong in thinking that P300 detection methods cannot be beaten. The P300 wave is a response that is produced when the stimulus is meaningful to the subject; it can therefore be manufactured by any method that makes irrelevant probes (used to set the baseline for comparison of wave amplitude) meaningful to them. Apparently, this is not very difficult. Rosenfeld and colleagues (Rosenfeld et al. 2004) instructed subjects to perform a variety of covert acts in response to irrelevant stimuli: wiggling toes, pressing the fingers of the left hand onto their legs and imagining the experimenter slapping them in the face. These coun- termeasures were sufficient to defeat Farwell’s six-probe paradigm. Reaction-time data remained a significant predictor of guilt on a one- probe variety of the test, but this is of little comfort to proponents of MERMER-based guilty knowledge tests; the six-probe test is neces- sary to avoid too great a rate of false positives as a consequence of subjects’ finding the probe meaningful for coincidental reasons. It is apparent that the P300 test has not, so far, lived up to the hype. However, it may well improve, as the hardware gets more sophisticated, the algorithms that interpret the data are refined and the experimenters find better ways to probe guilty knowledge. Moreover, the P300 test is only one of several deceit-detection technologies currently under investigation, some of which are also aimed at detecting lies by reading brains. Langleben and colleagues (Langleben et al. 2002) used fMRI to scan the brains of subjects engaged in intentional deceit; they discovered that areas of the anterior cingulate cortex and the superior frontal gyrus were more mind reading/controlling minds 137 active during deception than when subjects responded truthfully. Once again, there are questions concerning the ecological validity of the technique: how well will it generalize from the controlled laboratory with willing subjects to the outside world where condi- tions are uncontrolled and subjects are uncooperative? Even in the laboratory, the accuracy of the test is not all that high: it showed a between-group difference between deceivers and the truthful, but the effect is not great enough to identify individual deceivers with a high degree of confidence. In addition, technical limitations of fMRI – its relative lack of spatial and temporal resolution – will probably need to be overcome before the test has an acceptable degree of reliability. Once again, however, the technology and the testing methods can be expected to improve. What does the foreseeable future hold? I think it is safe to claim that the kind of mind reading technology which is most feared, which can scan the brains of subjects and reveal intimate details about their thoughts, without their knowing that they are under the mental microscope, is (at least) a long way off. The most promising methods of mind reading require that we build up a set of data on an individual subject: we need to establish a baseline for responses we know to be truthful, against which to compare the probes of interest (see Illes et al. 2006 for review). Conditions must be carefully con- trolled and the subject (relatively) cooperative. Moreover, neither EEG equipment nor, especially, fMRI equipment, is anywhere near portable or concealable. We can expect to see mind-reading tech- nology that is of some help in detecting deception in the laboratory, and that can therefore be used in the kinds of situations in which polygraphy is employed today, long before we see covert surveillance of thoughts – if indeed that ever turns out to be possible. What about laboratory tests for mental states and dispositions other than lies? Once again, work is proceeding along several fronts. Many studies have shown brain alterations associated with chronic schizophrenia; the possibility therefore exists that the disease could be diagnosed on the basis of brains scans (Farah and Wolpe 2004). reading minds/controlling minds 138 Some researchers claim to have discovered identifiable neural correlates of normal personality traits. The work of Canli and col- leagues (2001; 2002) is the best known and most interesting in this vein. They found that extraversion was correlated with particular kinds of responses to images with positive emotional qualities, whereas neuroticism was associated with differences in responses to images with negative emotional content. Phelps and colleagues (2000) used fMRI to study the responses of white subjects to photo- graphs of black faces. They found a correlation between the degree of activation of the amygdala and negative evaluation of blacks. We shall return to this study shortly. What are the prospects for a genuine mind-reading machine – one that is capable of interpreting brain states more generally? Out- side of personality traits, neuroscientists have had some success in detecting the neural correlates of the orientation of lines to which a subject is attending (Kamitani and Tong 2005); when subjects viewed a visually ambiguous figure, the researchers were able, on the basis of fMRI data, to determine how the subject was resolving the ambi- guity. Once again, the data was interpretable only after an initial ‘‘training’’ run was used to establish a baseline. Quiroga et al.(2005) claim to have been able to isolate neural correlates of a much wider range of representations. In their small study, they apparently showed that representations of a single person, building, or a single class of objects – e.g., cartoons from The Simpsons – are encoded in such a way that, no matter how they are presented, they activate specific neurons. Thus, one of their subjects had a neuron that responded preferentially to pictures of Jennifer Aniston, no matter what angle the picture was taken from, and relatively little to pic- tures of other people, famous or non-famous. Another subject had a neuron that responded preferentially to pictures of (what he took to be) the Sydney Opera House, as well as to the words ‘‘Sydney Opera,’’ but not to other buildings or people. This study used recordings of single neurons, rather than fMRI or EEG techniques. It might seem to provide the basis for a much mind reading/controlling minds 139 more powerful mind-reading technique, since the range of thoughts that it could detect is far wider than other techniques. However, there are good reasons to suggest that it will not result in a mind- reading machine anytime soon. First, the technology is invasive, requiring electrodes to be implanted deep in the brain (Quiroga and colleagues were only able to carry out the experiment because they had a pool of intractable epileptics, who required surgery, to draw on. Electrodes are implanted in the brains of such subjects to locate the foci of seizures, to allow the surgeon to locate the precise area for intervention). Second, once again the technique required training and cooperation. Even if we could get single neuron recordings from subjects, we could not interpret them unless we already had a set of data showing correlations between the firing of the relevant neurons and particular mental states. Third, the authors themselves caution that the fact that they were able, in the short time they had available, to discover pictures to which the individual neurons responded suggests that the neurons probably respond to other images as well. If the Jennifer Aniston neuron responded only to pictures of Jen, it would be nothing short of a miracle that the researchers had been able to hit upon its precise stimulus. But if the neurons respond to many different images – and perhaps to sensations and abstract thoughts as well – then even the possession of single-cell recordings plus a set of correlation data will not be sufficient to tell us what the subject is thinking. We may have to conclude that he is thinking about Jennifer Aniston, or parliamentary democracy, or Friday Night Football, or a pain in his toe, or something else for which we don’t have any data yet. The development of a general mind-reading technology, able to read the thoughts of people even in the absence of preliminary training and the establishment of a baseline, is possible only if there is a great deal of commonality in the neural correlates of mental states across persons. That is, it will only be possible to construct a device to read the thoughts of anyone – whether for the purposes of detecting potential terrorists or selling them cola – if it is possible to reading minds/controlling minds 140 construct some kind of translation manual, which details the correlations between particular brain states and particular thoughts. If my thought that ‘‘elephants are gray’’ has neural correlates which are very different from your thought that ‘‘elephants are gray’’, then constructing the translation manual will be difficult or impossible (impossible if there are few commonalities across the population; difficult if the differences are tractable – for instance, if there are identifiable groups, between which the neural correlates of thoughts differ markedly, but within which there is a great deal of common- ality – just as there is a great deal of commonality within, but not between, the vocabularies of different languages). Moreover, even the construction of a mind-reading machine reliably able to read the thoughts of a single person, upon whom the machine has been trained, depends upon our thoughts having stable neural correlates across time. Perhaps my thought that ‘‘elephants are gray’’ today has a very similar neural realization to the same thought, in my head at least, tomorrow, but perhaps next week, or next month, or next year, it will be quite different. There is already evidence for some kinds of commonalities within and across subjects. The method used by Kamitani and Tong (2005) to detect the orientation of lines to which a subject is attending uses data from extensive testing of the visual systems of monkeys; from this data, we know that orientation is represented in the early stages of visual processing in ways that are consistent across primates. However, the degree of consistency is not sufficient to underpin the development of a mind-reading machine: an initial set of data is necessary to make the neural activity meaningful. Thanks to the data on the primate visual system, we know where to look for orientation-tuned neurons, but gathering data on individual subjects remains indispensable for applying the technique. It’s not difficult, however, to think of ways in which this data could be gathered unobtrusively; that is, without the subjects’ being aware that it is taking place. We could flash lines in such a way that subjects had their attention attracted to them, and use this mind reading/controlling minds 141 information to hone in on the relevant neurons. Perhaps we could, but so what? This is, after all, a relatively uninteresting mental state: it is difficult to imagine a realistic scenario in which knowing the orientation of lines to which subjects are attending is important enough to justify the massive investment necessary to justify designing and building a machine able to provide such information. Moreover, even if such a machine were in existence, we would have little reason to worry about it. There are good reasons to worry about a loss of privacy, but few to worry about the loss of this kind of privacy in particular. Apart from a very few, very peculiar, situations, none of us will be worried if others are capable of determining the orientation of lines we are attending to. But mightn’t the results be capable of generalization; that is, mightn’t we expect to be able gradually to expand the range of mental states we are capable of detecting? The answer depends, once again, upon the degree of commonality in the neural correlates of thoughts across subjects and within subjects across time. The visual system, especially the early visual system, is relatively (the emphasis is very necessary here) simple, and the relationship between its contents and what it represents relatively easy to decode. It may turn out, for all we know at the moment, that more complex and abstract thoughts have neural correlates that are far less stable across time and across subjects. In recent work, Haynes and colleagues were able to identify whether a subject had chosen to add or to subtract two numbers with seventy percent reliability (Haynes et al. in press). Arithmetical operations are at a greater level of abstraction than visual perception, but this is still a long way off from complex conceptual thought. Indeed, such operations, and all the other kinds of mental states that we have hitherto been able to detect via fMRI, may be subserved by brain modules, innate brain structures dedicated to specific tasks. If an important cognitive task regularly confronted our distant ances- tors in the environment of evolutionary adaptation, then a module may have developed to perform it efficiently. Since brain modules are (typically) discrete entities, and perform a regular function in a reading minds/controlling minds 142 [...]... to get them to think, desire and believe what others want If methods of controlling minds neurotechnologically can be devised, our autonomy, our rightly prized 145 146 reading m inds/controlling minds ability to shape our lives according to values that we endorse, would be under threat Are covert means of controlling minds or behavior achievable, at least in principle? There is already evidence to suggest... (Ekman et al 1999) If we are worried about the ability of others to discover what is on our minds, it is not only neuroscientific technologies, and not even the techniques of cognitive and social psychology, that we must worry about; it is also the ordinary skills of 151 152 reading m inds/controlling minds mind reading that human beings possess, and which can be honed to a high degree of accuracy What... these responses need to be carefully rethought, in light of the fact that ‘‘mind -reading ’ is best done, today and for the foreseeable future, from outside the mind Generally, the best way to discover what someone is thinking is to ask them They are able to describe the contours of their 149 150 reading m inds/controlling minds thoughts with a detail and a subtlety, using that marvellous tool, language,... ‘‘pain,’’ ‘‘hurt,’’ ‘‘agony’’ – appears on the screen, and to press a button the right whenever a white face or a positively valenced word – ‘‘beautiful,’’ ‘‘pleasure,’’ ‘‘joy’’ – appears 147 148 reading m inds/controlling minds They then perform the second task, sorting words into ‘‘white and bad’’ or ‘‘black and good’’, again signalling category by pressing a button (of course, the order in which the two... depend on when I learned about her; in what context, both internal and external This context will certainly differ from person to person Useful, or threatening, mind -reading machines are not going to be built soon: not, at least, mind -reading machines capable of detecting the details of complex thoughts Machines that detect, say, emotional arousal are another matter: given that the structures for emotional... recalling a past event seems subtly to alter the memory, so that when the event is recalled a second time, it is the remade memory that is recalled Older memories are also altered 143 144 reading m inds/controlling minds by the context of recall, and by newer memories since laid down (Schacter 1996) It may be that a very large number of our mental states, especially our complex thoughts which represent... want to make someone buy a product that they desire, but which they prefer, all things considered, not to purchase, you ought to ensure that their self-control resources are depleted 153 154 reading m inds/controlling minds when they confront the option of purchasing it This can be done by requiring potential purchasers to engage in self-control tasks, which deplete their resources, before they are presented... about the second as well conclusion In this chapter, we have seen that there are no special reasons to worry about neuroscientific mind reading or mind control; the kinds of powers that neuroscience promises in the near future pale in conclusion comparison to the mind reading and control techniques already in existence, in power and in precision Autonomy is an important good, and we ought not to surrender... concerned especially about internal means of manipulating or reading the mind: not unless and until far more powerful techniques come into existence than currently seem practical Thus, the issues of brain privacy and mind control demonstrate the ethical parity principle in action: we see that the very same reasons we have to fear neuroscientific mind reading and mind control apply, with at least equal force,... what he takes to be ‘‘empirical evidence that data derived from brain scans can be better predictors of behavioral measures than other types of measures’’ (2006: 174) However, his 155 156 reading m inds/controlling minds evidence shows no such thing What he demonstrated is that fMRI data better accords with measures of behavior than personality questionnaires do Two points should be made about this . I leave this fascinat- ing topic for the next chapter. reading minds/ controlling minds 146 mind reading, mind controlling and the parity principle I mentioned. say that thoughts are realized in brains, so the claim should reading minds/ controlling minds 134 be uncontroversial to anyone who does not believe in