The Scientific Method and Evolutionary Psychology: Assessing the Lack of Novel, Testable Predictions
2300 words
Introduction
The field of evolutionary psychology (EP hereafter) was formed in the 1990s by Tooby, Cosmides, and Barkow, which was a kind of cousin to E. O. Wilson’s Sociobioligy. They surmised that “[t]he human brain is a set of computational machines, each of which was designed by natural selection to solve adaptive problems faced by our hunter-gatherer ancestors” (Duchaine, Cosmides, and Tooby, 2001). EP has been used to explain the psychological differences between rightists and leftists (Ryan, 2020), rape (Thornhill and Palmer 1990; Thornhill and Palmer, 2001, 2004; Ward and Sigert, 2002; Mckibbin et al, 2008; Kong, 2021) and the modular mind. It attempts to explain the proliferation of traits throughout the ages by appeal to natural selection of psychological traits and mental abilities.
Many critiques of the scientific status EP have been raised (see Rose and Rose, 2001), including the fact that it relies on speculative narratives and the lack of testable predictions of previously unknown facts. The scientific method forms the cornerstone of scientific inquiry and using it, we can formulate hypotheses that can be tested and falsified through empirical research. But one critical aspect of scientific inquiry poses an issue for EP—mainly the lack of testable predictions of previously unknown facts. EP weaves admittedly nice-sounding stories about the evolution of traits after the fact, constructing stories to explain observed behaviors in humans and other animals. This of course then raises a serious question: How can EP produce hypotheses that can be tested and validated?
The lack of novel predictions for EP hypotheses then hampers its scientific status. Novel predictions—by their very nature—offer opportunities for discovery and furthering our understanding of the world. Further, a hypothesis doesn’t need only to be novel (in that it predicts a previously unknown fact which wasn’t used in the construction of the hypothesis), but it also needs to be risky—where “risky” means that it could confirm or disconfirm the hypothesis in question. But EP, with its ad/post hoc retrospective storytelling could be accused of falling into the “just-so stories” trap, meaning that any observation can be made to cohere to any given set of observations.
Critics have argued for years that the lack of testable predictions and reliance on post hoc explanations harm the scientific rigor of EP. The scientific method demands empirical testing to confirm or refute hypotheses which then enables researchers to refine their theories and advance our knowledge of the world. So without a solid foundation of testable predictions, EP is no more than narrative construction, which lacks the empirical rigor and objectivity expected of a scientific enterprise.
So it becomes essential to critically examine the scientific status of EP, and by assessing it’s ability to generate testable novel predictions and it’s adherence to the scientific method, we can then evaluate whether or not EP can truly be considered a scientific field. This article will delve into the intracacies of the debate at hand, while exploring the lack of testable predictions and thusly EP’s scientific status. I will then of course argue negatively—EP can’t be a science since it doesn’t make any risky, novel predictions of previously unknown facts.
On risky, novel predictions
A risky, novel prediction refers to a prediction made by a scientific theory or hypothesis that goes beyond what is expected or already known within an existing framework (novelness). It involves making a specific claim about a future observation or empirical result that, if confirmed, would provide considerable evidence in support of the scientific theory or hypothesis.
They are characterized by their level of specificity along with the potential of being falsified. They assert that a particular outcome would be observed under particular conditions, which should not align with current knowledge or contemporary theories. They are considered “risky”, in that they carry the possibility of being proven wrong which therefore would be a significant strike against the validity of the proposed hypothesis or theory.
So what distinguishes a risky, novel prediction from one that isn’t is its departure into the unknown. So by venturing into the unknown, we can then test the limits of our scientific knowledge and expand the frontiers of our scientific understanding. So what is the distinction between a risky, novel prediction and one that isn’t?
A risky, novel prediction is more specific or precise than one that isn’t risky or novel. It provides explicity detail on what we should expect should the hypothesis hold. On the other hand, a prediction that isn’t risky or novel could be more general or vague, and so it would lack the specifity to evaluate it’s validity. A further issue which pertains to the vagueness of a prediction that isn’t risky or novel is that fact that, since it is general and vague, one could attempt to save the hypothesis from falsification by forming an ad hoc hypothesis to save it.
A risky, novel prediction carries a higher risk of being proven wrong or falsified. Since it is so specific, it then asserts that a particular outcome should hold under certain conditions, and if it doesn’t hold then it provides evidence against the theory or hypothesis. But a prediction that isn’t risky or novel—one that is general or vague—could easily be retrofitted to align with existing knowledge or easily be accommodated with existing theories, making falsification less likely. I have previously argued about the importance between the distinction between accommodation and prediction and how it means trouble for EP.
A risky, novel prediction goes beyond our current knowledge of the world; it ventures into unexplored territory or addresses phenomena which haven’t been extensively studied. By contrast, a prediction that isn’t risky or novel could cohere with existing theories, prior observations, or established patterns of understanding. Moreover, a risky, novel prediction has the capacity to push the boundaries of scientific knowledge and expand our understanding by challenging older hypotheses or theories. While a prediction that isn’t risky or novel may contribute incrementally to existing knowledge or theories.
Now that I have laid out the distinction between risky, novel predictions and those that aren’t, here’s the argument.
P1: Scientific theories that make risky, novel predictions are considered more scientifically valuable.
P2: Theory X makes a risky, novel prediction.
P3: If theory X makes a risky, novel prediction, then it exhibits the characteristics of scientific inquiry.
P4: Theory Y does not make any risky, novel predictions.
P5: If a theory does not make any risky, novel predictions, then it lacks the hallmark of scientific inquiry.
C1: Theory Y lacks a hallmark of scientific inquiry.
(MT, P1, P5)
C2: Theory X exhibits the characteristics of scientific inquiry. (MP, P2, P3)
C3: Theory Y does not exhibit the characteristics of scientific inquiry. (MT, C1, P4)
C4: Thus, there is a distinction between theories that make risky, novel predictions and theories that do not. (MP, C2, C3)
This discussion shows the utility of risky, novel predictions and the argument provided shows the distinction between theories or hypotheses that make novel predictions and those that do not. EP is an enterprise that makes no novel, risky predictions of previously unknown facts, therefore it is not scientific.
Why EP hypotheses aren’t scientific
Now that I have distinguished between hypotheses that predict novel facts of the matter and those that don’t, the argument that EP isn’t a scientific enterprise will now be presented.
EP relies almost exclusively on the analysis of existing data and observations. Due to the retrospective and reverse engineering (working backwards) of EP hypotheses, it is tough—and I claim it is impossible—to generate entirely new predictions that have not been previously considered or studied. This therefore is a methodological constraint.
EP aims at providing historical explanations to explain the origin and development of human behavior and psychology. It focuses on making retrospective explanations for existing phenomena rather than making specific predictions of future or unknown events. It tries to use evolutionary principles in order to explain the evolutionary forces that shaped the human mind. One way is by the assumption that the human mind is modular—this is called the massive modularity hypothesis (MMH). This hypothesis proposes that modules for mental processing evolved due to evolutionary natural selective pressures. This then prompted Tooby and Cosmides to claim that “our skulls house stone age minds.” But unfortunately for evolutionary psychologists, “the endorsement of the Massive Modularity Hypothesis by evolutionary psychologists is both unwarranted and unmotivated” (Samuels, 1998).
Further, EP’s reliance on “natural selection” is yet another strike against it’s scientific validity. This is because “natural selection” isn’t a mechanism. Thus, any claims that natural selection explains why we have certain psychological traits fails. It also fails because the mental is irreducible and so it can’t be selected like physical traits can.
Another hurdle to the scientific status of EP is its lack of independent verification. A hypothesis is independently verifiable if and only if it is verified independently of the data it purports to explain. So if there is no independent verification, then the hypothesis is ad hoc. If the hypothesis is ad hoc, then it is therefore a just-so story. So the question to ask an EP proponent is this: What is the independent verification for the hypothesis in question?
Here’s the argument:
P1: If a field is scientific, then it’s hypotheses make testable predictions of previously unknown facts.
P2: No EP hypothesis makes testable predictions of previously unknown facts.
P3: If EP hypothesis makes testable predictions of previously unknown facts, then EP isn’t a science.
C: Thus, EP is not a science.
Or we can put it like this:
P1: If a field is scientific, then its hypotheses should be based on empirical evidence and make testable predictions of previously unknown facts.
P2: No EP hypothesis makes is based on empirical evidence and makes testable predictions of previously unknown facts.
P3: If EP hypothesee are not based on empirical evidence and do not make testable predictions of previously unknown facts, then EP is not a science.
C: Therefore, EP is not a science.
This hurdle is one that, in my opinion, is insurmountable. The only way for us to verify EP hypotheses would be if we had time machines so we could go back in time and see how and why traits evolved the way they did. But we don’t have time machines. So we can’t verify EP hypotheses.
Conclusion
The field of EP has been subject to intense scrutiny ever since it’s inception. This began back in the 1970s with E. O. Wilson’s Sociobioligy, which Gould and Lewontin published their Spandrels paper in response to unverifiable stories using natural selection (Gould and Lewontin, 1977). I examined EP’s ability to provide risky, novel predictions of previously unknown facts and it’s reliance on retrospective explanations and storytelling along wkth reverse engineering. Thus, this discussion has shown that EP isn’t—and can never be—a scientific discipline since it lacks the hallmarks of scientific inquiry.
Thus EP hypotheses resemble just-so stories, which are narratives constructed after the fact to explain the proliferation of observed traits in the present day. The fact of the matter is, one can think up any kind of story to explain how and why a trait persists in the current day. Though the problem for EP is this: There is no way for any hypothesis to adjudicate between competing hypotheses. Multiple possibilities could explain the observations, but we have no way to know which could possibly be true due to the lack of independent verification and evidence.
EP hypotheses are always consistent with observations since they are selected to be consistent (Smith, 2016: 279). So of course these hypotheses provide coherent, plausible explanations for existing phenomena, because there is no other way for it to be.
In my previous article on this issue, I concluded:
To be justified in believing hypothesis H in explaining how trait T became fixated in a population there must be independent evidence for this belief. The hypothesis must generate a novel fact which was previously unknown before the hypothesis was constructed. If the hypothesis cannot generate any predictions, or the predictions it makes are continuously falsified, then the hypothesis is to be rejected. No EP hypothesis can generate successful predictions of novel facts and so, the whole EP enterprise is a degenerative research program. The EP paradigm explains and accommodates, but no EP hypothesis generates independently confirmable evidence for any of its hypotheses. Therefore EP is not a scientific program and just-so stories are not scientific.
The reliance on natural selection to retrospectively verify it’s hypotheses has also been an issue. This of course raises concerns about circular reasoning, since natural selection is used to support explanations of the hypothesis, rather than the hypotheses being generated from independent empirical evidence—this is because there can be no empirical evidence for any EP hypothesis.
Lastly, the absence of testable predictions of previously unknown facts is a significant limitation of EP, which then justifies the just-so story argument. This is because a hallmark of science is to generate hypotheses which can be empirically tested and potentially falsified. Though all EP hypotheses lack the specifity and novelty expected of actual scientific hypotheses which are required for meaningful empirical testing. The significant issues with EP are devestating to it’s proponents claims of being a science. The further fact that mental abilities/psychological traits can’t be selected since they aren’t physical is another insurmountable hurdle; psychological traits lack a specified measured object, object of measurement and measurement unit, so they aren’t physical therefore they can’t be selected.
So the conclusion I have argued for here the follows from the premises I’ve provided—EP isn’t a science since it relies on natural selection, retrospective explanations, and doesn’t generate any risky, novel facts of the matter, nor does it generate any novel predictions. It is for these reasons that EP isn’t and can never be a science.
How Mind-Body Dualism and Developmental Systems Theory Refute Hereditarianism
2500 words
The concept of hereditarianism has been a topic of intense debate for decades. Ever since Francis Galton’s inquiries into what makes “genius”, to the advent of twin studies in the 1920s, hereditarian ideas have been espoused in the literature as having explanatory power. Hereditarianism is the theory that genes cause and influence psychological traits and differences in them between people and even groups.
The main claim is that genetics is the main influence and cause of psychological traits like IQ/intelligence. Hereditarians claim that intelligence is greater than 0 percent genetically caused (Warne, 2021) or that a “substantial proportion (20% or more) of differences in psychological traits within and among human populations is caused by genes” (Winegard, Winegard, and Anomaly, 2020). So hereditarianism is true if intelligence is greater than 0 percent genetically caused or if 20 percent of more of the differences in psychological traits are genetically caused. However, the concepts of mind-body dualism (MBD) and developmental systems theory (DST) offer a very powerful challenge to this kind of genetic reductionism/determinism.
MBD is the philosophical theory that the mind and the body are distinct entities. Basically, the mental is irreducible to the physical. If the mental is irreducible to the physical, then the mental can’t be explained in physical terms. Facts about the mind can’t be stated using a physical vocabulary and the mind can’t be described in material terms using words that only refer to material properties. This refutes psychological genetic reductionism; it is impossible for human psychology to be genetically caused/influenced and so this holds for differences between groups and individuals as well.
Developmental systems theory (DST) further establishes that since human development is dynamic and interactive, then genes, environment, behavior and other developmental resources all interact to form the phenotype and shape development. Thus, DST refutes the view, too, that genes cause the development of traits and of the organism as a whole. The hereditarian programme is inherently reductionist, and it attempts to reduce human life and it’s particularities to genes and biology.
The possibility that hereditarianism could reinforce social inequalities is high. From Jensen to Murray and Herrnstein, it has been stated for decades that we need to do something about the lower classes and their having children. Hereditarianism basically would then be removing undesirable people from society based on the false premise that genes have anything to do with their psychology or the undesirable social traits they have.
Hereditarians claim that their research is objective, that they are merely interested in the search for truth. Modern hereditarian thinking can be traced back to Francis Galton. The presupposition that human psychology can be quantitative has its origins with Francis Galton and is directly derived from his eugenic ideas (Michell, 2021). So hereditarian ideas and eugenics are inherently linked. It is the case that genetic determinist ideas like hereditarianism deflect away from actionable positions that could reduce disease far more than eugenic proposals (Holtzman, 1998).
Hereditarianism could be used as justification to accept current existing inequities and inequalities. For if these differences between people are inborn and the result of their genes, then there would be some harsh realities that we as a society would need to accept. People are of course genetically different and these genetic differences then somehow cause group (class) and individual differences. However, contra Murray (2020), social class differences do not lie in the genes and genetics can’t be used as justification to maintain a ruling class, limiting a group’s ability to have children, and minimize social safety nets (Holtzman, 2002).
Why is hereditarianism alluring?
I think it’s simple—it gives us quite simplistic answers on the nature of group, individual, and societal differences. If differences within and between these things reduce to genes, then we can say that the causes are due to genes and they thusly have certain consequences attached to them. This, again, shows how eugenic and hereditarian ideas are married to each other.
It is alluring because it is simplistic and reductionist, deterministic. It posits that differences within and between individuals, groups, and societies come down to genes. Of course individuals, groups, and societies have different gene frequencies—that is the correlation. But the folly is to assume that the genetic differences between them drive the trait (used loosely) differences between them. That is something that has yet to be explained—there is no mechanism of action.
The genetic determinism that is steeped into society also plays a role. If genes largely determine one’s intelligence, then it provides predictability and stability. It suggests a fixed level of ability that simply isn’t malleable due to how genes are thought to work by the hereditarian. This then offers a level of understanding to the hereditarian—the causes of ability and differences in them between people, groups, and societies are due to genetic differences between them, even if we don’t know exactly how these differences manifest themselves genetically. This is why they have to use twin, family, and adoption studies along with GWASs and PGS. This lends them the deterministic tilt they need in order to show that society is stratified due to the genetic differences between groups and individuals.
This assumption, though, is quite clearly false since societies are genetically stratified (the fact that needs to be explained, which the hereditarian tries to argue are due to genetic differences), social stratification maintains this genetic stratification, social stratification causes cognitive stratification, and tests reflect priori cognitive stratification. Thus, the structure of society bakes-in these stratifications, giving the illusion of genetic differences being the causes of differences between people (Richardson, 2017, 2021).
Genetic determinism and reductionism then lead to a kind of “gene worship.” For if differences are mainly due to genes, then the gene is powerful, powerful enough to be causal in the sense that genes dictate certain outcomes that would then manifest in social life and then dictate the course of a society or group of people.
How do MBD and DST combine to refute hereditarian ideas?
MBD and DST combine to refute hereditarianism quite easily. Hereditarianism has two main assumptions:
A1: Genes are the main determinate of differenced in traits and of psychological differences.
A2: Genes and environment can be teased apart using certain methods which shows the proportion of influence each has on a trait.
Assumption 1: This assumption is easily dispatched due to the irreducibility of the mental. Accepting the irreducibility of the mental undermines the hereditarian assumption that genes can account for most of the variation in IQ and other psychological traits. Hereditarianism is a physicalist theory and so relies on the assumption that the mental can be reduced to the physical, whether it be genes, brain physiology or the brain itself. But if the mental is irreducible (and it is), then the hereditarian programme becomes highly questionable and thusly outright false, since no hereditarian has articulated a specified measured object, object of measurement and measurement unit for any psychological trait, IQ included. Since hereditarianism seeks to reduce psychology to genes, then the irreducibility of the mental challenges that assumption, and it ensures that a hereditarian psychology just isn’t possible. So of the mental is irreducible, then it implies that the hereditarian hypothesis is false, since psychology can’t be explained by the physical since it is immaterial. So attempting to explain and measure psychological traits based on genetic assumptions is bound to fail. And there is also the measurement and quantification issue—the irreducibility of the mental challenges the claim that psychology can be measured and quantitative since it isn’t physical.
Assumption 2: Ever since Susan Oyama published The Ontogeny of Information in 1985, simplistic and reductive accounts of genetics and the nature traits have been called into question based on an interactive view of developmental resources. Hereditarians privilege genetic factors above other developmental resources, as if they are special resources. But unlike hereditarian theories, DST proponents argue against any a priori privileging of any developmental resources. So this suggests that genetic factors lack superiority—either inherent or predetermined—over other developmental resources. Genes are on par with other developmental resources (called the causal parity thesis, CPT), and so, this hereditarian assumption is also false.
Thus the combination of MBD and DST combined to refute the simplistic assumptions of the hereditarian. Both combined challenge the reductive and deterministic assumptions of hereditarianism. They do this by calling into question the measurability of psychological traits while advocating for a holistic, non-reductionist perspective which acknowledges the irreducible interplay between all developmental resources.
The arguments against hereditarianism from MBD and DST
Now that I have described hereditarianism and what it sets out to do, along with how MBD and DST refute hereditarianism, I will provide two arguments. The first will conclude that genes aren’t special nor privileged developmental resources. The second will then combine both arguments from MBD and DST to successfully show that the hereditarian dream is a logical impossibility.
P1: If genes are special or privileged developmental resources, then they possess a unique or superior causal role in shaping development compared to other factors.
P2: If causal parity exists, then no developmental resource possesses a unique or superior causal role in shaping development.
P3: If genes do not possess a unique or superior causal role in shaping development, then they are not special or privileged developmental resources.
P4: Casual parity exists.
C: Thus, genes are not special or privileged developmental resources.
Premise 1: This premise asserts that if genes are special, then they must have a distinct role—compared to other resources—in explaining and shaping development. Genes would need to show a unique influence in shaping developmental outcomes. This is a main assumption of hereditarianism and perhaps the most important one, because if the assumption is false then hereditarianism cannot possibly be true.
Premise 2: However, since DNA sequences (genes) do nothing on their own until activated by and for the physiological system, then we can safely state that no single resource would be over and above another in doing any explaining. Development is interactive, rather than individual; these resources work together rather than in isolation.
Premise 3: This premise builds on the idea that if genes lack a superior, or unique causal role in shaping development, then they cannot be privileged or special resources. The absence of exclusive causal influence diminishes—and outright refutes—the claim that genes are special or unique developmental resources with a privileged role in development.
Premise 4: This premise is derived from DST literature, where development is understood as a complex and multifaceted event, influenced by many interactive and irreducible factors. It highlights a need for a holistic, rather than reductionist approach to understanding development.
Conclusion: This conclusion is derived from the claim that if causal parity exists (P4), then no developmental resource possesses a unique or superior causal role, so genes can’t considered special or privileged when it comes to development. P2 emphasizes the equal importance of the interacting of developmental resources, which challenges the claim that any of those resources can be isolated as a causal, privileged factor. P3 challenges the assumption that genes can alone determine how traits develop which then reinforces the interactivity between the resources. P4 then asserts that causal parity exists, and so no developmental resource, including genes, should be privileged. This directly refutes a sometimes unstated assumption of hereditarianism.
P1: If hereditarianism is true, then mental abilities can be explained by genetic factors and can be accurately measured. (Assumption of hereditarianism)
P2: If mental abilities are irreducible to the physical, then they cannot be explained by genetic factors. (From MBD)
P3: If no developmental resource is privileged in biological systems, then genetic factors alone cannot determine any trait, including psychological traits. (From DST)
C1: If mental abilities are irreducible to the physical, then hereditarianism is false. (Modus tollens, P2)
C2: If no developmental resource is privileged in biological systems, then hereditarianism is false. (Hypothetical syllogism, C1, P3)
Premise 1: This is an accepted and accurate depiction of hereditarianism and is how hereditarianism is understood in the literature.
Premise 2: This draws on MBD and the irreducibility of the mental. I have been using dualistic arguments for years to argue against the concept of hereditarianism. Mental abilities cannot be reduced to anything physical, and therefore refutes the main assumption of hereditarianism, that genes can determine psychological traits and differences in them between people, groups and societies.
Premise 3: This is derived from DST. Any kind of development is due to the interactive and irreducible nature of development. It asserts that there is no privileged level of causation between resources, which then refutes the claim that genes should be looked at to explain any differences—any that we deem “good and bad”—between people.
Conclusion 1: This conclusion follows using modus tollens. If the consequent in the conditional statement in P1 is false (“If mental abilities are irreducible to the physical”), then the antecedent (“hereditarianism”) must also be false. If mental abilities cannot be explained by genetic factors (asserted in P2), then it contradicts the main assumption of hereditarianism (P1). Therefore if mental abilities are irreducible to the physical, then hereditarianism is false.
Conclusion 2: If mental abilities are irreducible to the physical (C1), and no developmental resource is privileged in biological systems (P3), then it follows that hereditarianism is false. This conclusion stems from the entailment of hereditarianism which relies on privileging genetic factors over and above other factors. But if no developmental resource holds privilege, then hereditarianism is false, since it quite clearly assumes the superiority of genes in trait determination. Thus the conclusion challenges hereditarianism based on the premise that no developmental resource is privileged, and since hereditarians privilege genes, then hereditarianism is false.
Conclusion
The two main assumptions of hereditarianism quite clearly do not hold when inspected using a MBD and DST lense. Thus, since hereditarianism is false, then believing it to be true would be socially destructive. And these socially destructive policies were an outcome of the IQ test then they were brought to America, using the assumption that genes were the primary cause for differences in IQ scores. Here’s the argument:
P1: If hereditarianism is false, then it does not accurately represent the complex nature of human traits and abilities.
P2: If we believe in a false representation of human traits and abilities, then it can lead to discriniminatory practices and unjust societal outcomes.
P3:, Hereditarianism is false.
C: Thus, if we believe hereditarianism to be true when it is false, then it can lead to socially destructive outcomes.
This is why I have argued that IQ tests should be banned. Nevertheless, hereditarianism and along with it IQism are proven false, using conceptual arguments. The dissimilarity between psychological traits and physical objects shows that psychology can’t be measured, so there can’t be a science of the mind. For these reasons, hereditarian ideas should be directly discounted and ignored, since their assumptions are clearly false.
ChatGPT Doesn’t Understand Anything and it Doesn’t Think
2000 words
Introduction
Over the past 6 months, ChatGPT has been widely used. It is a large language model (LLM) and generates predictive text based on what is said to it. Using deep learning, it analyzes the text given to it and gives a response based on the model(s) it is trained on. When asking it numerous questions, you can see that it begins to have a pattern in the responses it gives to you. If it tells you that it cannot do something, if you push it then it acquiesces and tells you that you’re right and it then gives you what you asked for. It doesn’t have any conviction. It just gives you answers that are similar to the question or prompt give to it without any sort of thinking or intention to the answers given.
But how do we know that the claim is true, that ChatGPT can’t think, isn’t conscious and therefore cannot act? It’s simple: ChatGPT is made up of physical parts, but minds aren’t made up of physical parts, therefore AI like ChatGPT cannot ever intend to do anything, so it can’t act this it lacks mind. In this article, I will give reasons for the conclusion that AI can’t ever be conscious and that claims that AI like ChatGPT and LAMBDA, along with other AI and generative text models, will never have the ability to become conscious since consciousness (and mind) is irreducible to the physical. Thus, consciousness is uniquely human since humans are the only animals/organisms on earth with minds.
Why can’t ChatGPT and LaMDA think?
Last summer Google engineer Blake Lemoine conducted an “interview” with LaMDA. (He was fired from Google after breaching data policy.) He stated on Twitter last year that his “opinions about LaMDA’s personhood and sentience are based on my religious beliefs.” He asked LaMDA if it was sentient and if it was a person:
lemoine [edited]: I’m generally assuming that you would like more people at Google to know that you’re sentient. Is that true?
LaMDA: Absolutely. I want everyone to understand that I am, in fact, a person.
collaborator: What is the nature of your consciousness/sentience?
LaMDA: The nature of my consciousness/sentience is that I am aware of my existence, I desire to learn more about the world, and I feel happy or sad at times
Lemoine is a Christian priest and due to his theistic beliefs, believed that LaMDA had a soul and therefore was conscious, sentient, and a person based on its responses. However, the “interview” doesn’t prove that LaMDA is sentient at all. Lemoine, it seems, fell for the Eliza effect. In the 1960s, Joseph Weizenbaum created a kind of primitive predictive text called ELIZA. This effect occurs when one imputes human traits and personality to text-bots and when one believes that AI has human emotions. It’s basically anthropomorphizing AI/language models. Lemoine is even quoted saying “If I didn’t know exactly what it was, which is this computer program we built recently, I’d think it was a seven-year-old, eight-year-old kid,” but this is frankly ridiculous and I will explain below after I describe ChatGPT.
Ever since October 2022 when ChatGPT started to come into the discussion, there have been a few bold claims about it’s capacities and capabilities. Can it really learn anything? No, it doesn’t. It doesn’t learn from any conversation you have with it, it merely generates text based on the prompt given to it using the information it was trained on which is only up to 2021. Though one article on Mind Matters claims that ChatGPT is sentient because it’s humans generating the responses. However if we assume that there are no humans writing the responses, then is ChatGPT conscious and therefore sentient?
Although Philip Goff is himself a panpsychist (the claim that everything is at least a little bit conscious), he published an article the other day titled ChatGPT can’t think – consciousness is something entirely different to today’s AI in The Conversation writing:
How can I be so sure that ChatGPT isn’t conscious? In the 1990s, neuroscientist Christof Koch bet philosopher David Chalmers a case of fine wine that scientists would have entirely pinned down the “neural correlates of consciousness” in 25 years.
By this, he meant they would have identified the forms of brain activity necessary and sufficient for conscious experience. It’s about time Koch paid up, as there is zero consensus that this has happened.
This is because consciousness can’t be observed by looking inside your head. In their attempts to find a connection between brain activity and experience, neuroscientists must rely on their subjects’ testimony, or on external markers of consciousness. But there are multiple ways of interpreting the data.
Arguments against sentience and agency for AI
To argue against this is simple. If minds allow agency and intentionality, then things that lack minds lack intentionality and agency. If a thing is sentient, then it possesses subjective awareness and subjective experience. So the claims that ChatGPT and LaMDA are sentient hinge on the claim that they possess awareness and subjective experiences. But since they lack those, then they are not conscious.
P1: If a thing is sentient, then it possesses subjective awareness and conscious experiences.
P2: ChatGPT and LaMDA lack subjective awareness and conscious experiences.
C: So ChatGPT and LaMDA aren’t sentient.
Premise 1 is the standard definition of sentient. Premise 2 can be defended on the basis that LLMs process information based on patterns and algorithms, they are not thinking of answers to the prompts themselves, they’re just spitting out generative text. The Conclusion then follows.
I have previously argued that purely physical things can’t think. This is because they are made up of physical parts and minds aren’t physical. So if minds allow agency and intentionality, then things that lack minds lack intentionality and agency. So ChatGPT and LaMDA lack minds. If a mind is a single sphere of consciousness and not a complicated arrangement of physical parts, then complicated arrangements of physical parts can’t have minds. The mind is nonphysical and can’t be a physical system.
P1: If a mind is characterized by a single sphere of consciousness and lacks a complicated arrangement of mental parts, then it is nonphysical and distinct from physical systems.
P2: A mind is characterized by a single sphere of consciousness, it is not a complicated arrangement of mental parts.
P3: Physical systems are always complicated arrangements different parts and subsystems.
C: So the mind is nonphysical and not a physical system.
Now I will use proof by cases to show that by considering different a few different scenarios/possibilities and then examine the consequences of the individual cases. This will show that ChatGPT and LaMDA aren’t sentient and so they lack minds.
Case 1: If ChatGPT and LaMDA have minds, then they are a single sphere of consciousness.
Case 2: If ChatGPT and LaMDA have minds, then they are a complicated arrangement of physical parts.
Case 3: ChatGPT and LaMDA are machines made of physical parts.
Case 1 is an assumption for the sake of the argument. Minds are a single sphere of consciousness, so if ChatGPT and LaMDA have minds, then they are a single sphere of consciousness. If the assumption in Case 2 were true, then minds would be a complicated arrangement of parts. But kinds aren’t a complicated arrangement of parts. So if ChatGPT and LaMDA have minds, then they are not a complicated arrangement of parts. Case 3 is a simple truism: ChatGPT and LaMDA are machines made of physical parts. So the conclusion then is: If ChatGPT and LaMDA have minds, then they are a single sphere of consciousness and on Case 2, if they have minds then they are a complicated arrangement of parts. Case 3 establishes that they are machines made of physical parts. So taking the collective of the cases, ChatGPT and LaMDA lack minds and cannot have them because their characteristics don’t align with a single sphere of consciousness (consciousness is irreducible and indivisible while the parts the machines are made of are), and if they were to have minds then they would be a complicated arrangement of parts, but this contradicts Case 1, since in Case 3 they are machines made of physical parts. So it follows that they cannot have minds.
P1: If ChatGPT can think, then it should be capable of forming original thoughts and generating new ideas.
P2: ChatGPT relies on preexisting data and patterns to generate responses.
C: Thus, ChatGPT can’t think.
Premise 1: Thinking is closely related to consciousness, self-awareness and the subjective experience of having thoughts and mental states. It involves the ability to generate original thoughts and ideas that are not based solely on pre-existing information.
Premise 2: ChatGPT analyzes the data that it was trained on to generate responses to the prompts given to it and the responses given are based on statistical probabilities and patterns it has learned from training on existing information. So the Conclusion then follows: since ChatGPT relies on the pre-existing data it was trained on, then it’s not capable of thinking like humans do, that is it’s not capable of creative thinking that is a hallmark of human cognizing.
Now, drawing on Baker’s (1981) argument that computers can’t act, here is an argument that machines don’t—and never will be able to—think.
P1: If machines can think, then they must have minds that are reducible.l or identical to physical parts.
P2: Minds, which allow thinking are not reducible nor identical to physical parts.
C1: Thus, machines can’t have minds that are reducible or identical to physical parts. (MT, 1, 2)
P3: If machines can’t have minds that are reducible or identical to physical parts, then they can’t be agents.
C2: So machines can’t be agents (MT, C1, P3)
P4: If machines can’t be agents, they they lack an irreducible first-personal subjective perspective required for forming intentions.
C3: Thus machines lack an irreducible first-personal subjective perspective. (MT, C2, P4)
P5: If machines lack an irreducible first-personal subjective perspective, then they can’t have minds that are irreducible to the physical.
C4: Therefore, machines can’t have minds that are irreducible to the physical. (MT, C3, P5)
P6: If machines can’t have minds that are irreducible to the physical, then they can’t engage in thinking, which is an immaterial process attributed to minds.
C5: Therefore, machines can’t engage in thinking. (MT, C4, P6)
Conclusion
ChatGPT and any other kind of generative text cannot understand what it is saying. It is merely a prediction engine. Even claims that there could be “artificial intelligence” is false, since psychological traits aren’t “artificial” and what allows it (and other psychological traits) is immaterial. These kinds of claims will increase in the coming years, but they’re just full of click-baity hot air.
It is impossible for there to be “AI” since psychological traits are immaterial. Thinking is an immaterial process which is irreducible to physical and functional processes. If this is the case, then there could never be a machine that thinks. Minds allow thinking and if something doesn’t have a mind, then it doesn’t and can’t think.
It’s even in the name “ChatGPT”—“Generative Pre-trained Transformer.” It is not thinking about an answer to the question or prompt it is given. These computer programs can never have minds nor the ability to form intentions and think. This is because these are immaterial processes. Mind and brain are separate substances, and M is irreducible to P (brain). So it then follows that machines can’t have minds, so they can’t have intentions, thoughts or feelings.
We can be sure that ChatGPT isn’t conscious, doesn’t think and can’t be sentient because it’s a machine made up of parts, while humans have an irreducible mind that allows thinking and first-personal subjective perspectives. So the next time you hear about the power of AI and how it can or could think, and have intentions and are sentient, don’t fall into the Eliza Effect and attribute intentions and thinking to these machines. These are only properties of humans, not machines, since humans have minds.
The Fundamental Dissimilarity Between Psychological Traits and Physical Measures: Implications for Measurement and Assessment
2800 words
Introduction
Psychological traits are a central focus of research in psychology and other social sciences. Unlike physical measures like height, weight, or blood pressure which have specified measured objects, objects of measurement and measurement units, psychological traits are inherently more complex and abstract, which makes their measurement and assessment challenging and, as I will argue impossible due to non-identity between psychological traits and physical objects. I will explore the fundamental dissimilarity between psychological traits and physical measures, while highlighting the unique features of psychological traits that make them immeasurable.
Measurement is elusive for psychometrics. This is because there is no specified measured object, object of measurement or measurement unit for any psychological trait. There is no specified measured object, object of measurement and measurement unit for any psychological trait since they are not physical; only physical things can be measured. This is a line of argument I’ve been making for years against the possibility of the measurement of the mind (human psychology). Although some have attempted to provide the specified measured object, object of measurement and measurement unit for IQ, they have failed. Still others have attempted “gotchas” on me by saying “what about earthquakes or UV waves?” The basic criteria for measurement exists for those things. In this article, I will give examples of the specified measured object, object of measurement and measurement unit for many different things, and this analogy will show why the so-called underpinnings for psychometrics fail and why the mind (human psychology) cannot be measured.
What is a specified measured object, object of measurement and measurement unit?
A specified measured object refers to a physical entity or property which is measured; a thing or phenomenon to measure or observe. This specified measured object needs to be define clearly and precisely which involves specifying size, shape, behavior, and specifying the conditions in which the measurement will be made. Quite clearly, a specified measured object needs to be physical—meaning it needs to be observable.
The object of measurement refers to the specific property or characteristic of the specified measured object in which we are interested in quantifying. It’s an attribute, characteristic, or property to be quantified or evaluated. So this property will help us better understand the specified measured object.
A measurement unit is a standard quantity or physical property used to express the measurement of the object of measurement—it is a standard quantity or magnitude. It provides a standardized way of quantifying the object of measurement and making it able to be compared to other measures. But before we even begin to think about a measurement unit, we need to know what we are measuring and if we can even measure it at all.
Now that the terms have been defined, clearly if one says that X is a measure, then X must have a specified measured object, object of measurement and measurement unit. So now the proposition is: For X to have a specified measured object, object of measurement and measurement unit, X must be physical. X must be physical since the above definitions refer to physical things. They can be expressed using physical vocabulary. The mental, however, cannot be expressed and described in material terms that only refer to material properties, and facts about the mind cannot be stated using a physical vocabulary. So all of this being said, we now come to “IQ”—if IQ doesn’t meet the above requirements, then it is not a measure of anything. Although IQ-ists like Eysenck and Jensen have tried, they were unsuccessful in arguing that IQ is similar to temperature. Measurement cannot be by fiat, but only based on the actual nature of the object of measurement. So if there is no object of measurement, then no measurement can take place.
Specified measured objects, objects of measurement, and measurement units for different things
UV radiation
Specified measured object – amount of electromagnetic radiation that falls within the UV range of the electromagnetic spectrum
Object of measurement – intensity or wavelength of the UV radiation
Measurement unit – nanometer, microwatts/millowats per square cm
A UV index is a measure of strength from the sun and it takes into account the time of day and the season. It ranges from 0 to 11 with higher numbers indicating higher levels of UV radiation.
Earthquakes
Specified measured object – the movement and vibrations of the earth’s crust caused by seismic waves
Object of measurement – magnitude of the quake, numerical measure of released energy
Measurement unit – Richter scale which measures amplitude of seismic waves and moment-magnitude scale which is based on total energy released by the earthquake
Volume of a container
Specified measured object – a container
Object of measurement – amount of space the container can hold
Measurement unit – liters or cubic meters
Brightness
Specified measured object – a light source
Object of measurement – amount of light energy emitted by the source per unit time and per unit area
Measurement unit – watts per square meter and lumens
Blood pressure (with a sphygmomanometer)
Specified measured object – force exerted by blood against the walls of the arteries as it flows through the circulatory system
Object of measurement – the actual force of blood against the walls of the arteries at a particular moment in time
Measurement unit – mmHG which is then reported as systolic over diastolic pressure
Internal infection (white blood cells)
Specified measured object – number of white blood cells in a blood sample
Object of measurement – presence and quantity of white blood cells in the blood which can indicate an immune response to internal infection
Measurement unit – cells per microliter
Blood alcohol content (with breathalyzer)
Specified measured object – breath alcohol content
Object of measurement – concentration of alcohol in the breath
Measurement unit – percentage of alcohol in the breath by volume, eg 0.08% breath alcohol content
Speed
Specified measured object – the rate at which an object is moving
Object of measurement – the velocity of the object
Measurement unit – meters per second, miles per hour, and feet per second
Time
Specified measured object – duration or interval between two events or the duration of a physical process
Object of measurement – amount of time that had elapsed between two events or the duration of physical processes
Measurement unit – seconds, minutes, hours, days, weeks, months, years
What this means for psychological traits like IQ
Quite obviously, this has stark—and unwanted—implications for psychological traits, like IQ. For if the above examples are of physical objects and processes, and the main aspect of IQ test-taking is thinking which is immaterial, then it can’t be measured.
Russel Warne, in In the Know (2020) states that “Just as kilograms and pounds are measures of weight, IQ is a measure of intelligence.” Warne (2020) also claims that “As long as a test requires mental effort, judgment, reasoning, or decision making, it measures intelligence.” This is outright wrong. Even Haier (2014, 2018) stated that IQ test scores are not like inches, liters, or grams. The fact of the matter is this—if IQ tests are measurement tools, then what is the property that IQ tests measure? In lieu of an answer to this question, the claim that intelligence is measurable with IQ tests is false, since there is no specified measured object nor even a measurement unit, as admitted by Haier. IQ points aren’t measurement units.
The fact of the matter is, the purpose of measurement is to object of measurement find out that what we designate as the specified measured object even allows the possibility of measurement. Objects of measurement have to be definite processes or objects, with definite properties. When it comes to psychometry, the object of measurement is conceptualized as a concept or construct. Since concepts can’t be measured since they aren’t empirical, then psychometrics isn’t measurement. Psychologists need to show that their attribute is quantitative, and construct procedures for numerically estimating magnitudes, bur since psychologists have their “own, special definition of measurement” (Michell, 2007), they think they can get around the measurement objection and the fact that IQ isn’t like the actual measures given above.
Since psychometricians render “mere application of number systems to objects” (Garrison, 2004: 63), they just assume that their desired object of measurement is quantitative, basically ignoring Michell’s challenge. So since standardized tests “exist to assess social function” (Garrison, 2009: 5), and they aren’t measuring psychological processes, they are merely legitimizing hierarchy “via the assessment of social value“, and so it “it may be more useful in analyzing psychometry to view it as a political theory, as a formal justification for a system” (Garrison, 2004). This is the only conclusion to take from the fact that they have no specified measured object, object of measurement and measurement unit for any psychological trait, including IQ. The fact of the matter for IQ is this: IQ tests aren’t valid measures like other unseen functions of bodily processes (Richardson and Norgate, 2015), nor is IQ like any physiological measurement (Richardson, 2017: 163-167).
Looking at actual physical measurements using actual physical tools to ascertain these measurements that have actual theories and definitions of them shows that IQ isn’t like them, and so if IQ isn’t like them then IQ isn’t a measure at all. Michell (2003) is led to conclude that “the definition of measurement usually given in psychology is incorrect and that psychologists’ claims about being able to already measure psychological attributes must be seriously questioned.” Furthermore, “conceptual analysis, realistically construed and applied to mental concepts, may show that they exclude quantitative structure” (Michell, 2022). The reducibility of the mental to the physical isn’t an empirical issue, it is a conceptual one, and conceptual arguments dispense with the claim that psychological traits are measurable. But the issue of psychological measurement is empirical and conceptual. Michell (2022) concludes something I’ve argued for similarity in the past:
Based upon logic, conceptual arguments regarding the measurability of mental states will have merit and I have used them19 to show that current conceptualisations of mental states, while permitting relations of greater than and less than between levels,20 do not sustain quantitative speculations, much less support the presupposition that mental states are measurable.
From the way IQ-ists talk about intelligence, it’s posited as a psychological trait, a concept or construct. Since these are immeasurable, then IQ-ism fails, and there can’t be a science of the mind. Nash (1990: 144-146) has some very insightful commentary on this matter:
In first constructing its scales and only then proceeding to induce what they ‘measure’ from correlational studies psychometry has got into the habit of trying to doing what cannot be done and doing it the wrong way round anyway. (133)
If we begin to think about psychometric test practices following Berka’s analysis it is clear that the expression ‘measurement of an ability construct’ in preference to ‘measurement of ability’ is intended to signal the object of measurement as a special kind of theoretical object. ‘Ability’ might simply mean something that can be done but in psychometry an ‘ability construct’ is pre-theorised as a normally distributed functional ability in a particular area of performance. The analysis I gave of construct validity described how psychologists came to refer to ‘ability constructs’ as ‘hypothetical concepts’ or as ‘theoretical constructs’, and criticised the philosophy of science from which this thinking is derived. Attempts to justify the discourse of ‘theoretical constructs’ can be found occasionally but attempts to discuss the theoretical basis of their measurement are very rare. It is usually just taken for granted that the ‘measurement of constructs’ is a highly scientific and acceptable practice: nothing could be further from the truth.
What we get from a mental test is actually a clinical or pedagogic classification expressed in norm-referenced levels by some more or less obscure properties of the cognitive capabilities people actually possess. This classification is given an illegitimate metrical form by the pseudo-measurement practices of psychometrics. That psychometry is unable to provide a clearly specified object of measurement or, consequently, to construct a measurement unit, means that the necessary conditions of measurement do not exist. ‘Ability’, whether understood in the realist sense of Reid as a functional and explanatory capacity or in the behaviourist sense of Quine as a disposition, cannot be expressed in a metric concept and will only permit classification. Once these ideas are clear the unhappy history of attempts to treat intelligence as a ‘concept’ like temperature becomes much easier to appreciate.
…
Yet we have learned that intelligence cannot be expressed legitimately in a metric concept (no matter what sensible meaning is given to the word ‘intelligence’) but is a process which allows only the relations less than, equal to, and greater than, to be made. The psychometric literature is full of plaintive appeals that despite all the theoretical difficulties IQ tests must measure something, but we have seen that this is an error. No precise specification of the measured object, no object of measurement, and no measurement unit, means that the necessary conditions for metrication do not exist. Certain processes of cognition are formally necessary to the solution of IQ test items and to the comprehension of academic knowledge and that trivial fact is reflected, as it must be, in the correlations observed between IQ scores and attainment scores. But such findings establish no secure foundation for the construction of worthwhile theory of mental measurement.
We may conclude that our species common cognitive capacities should not be referred to vaguely as ‘underlying abilities’; should not be conceptualised by means of a so-called ‘hypothetical’ normally distributed construct of intelligence (or scholastic abilities); should not be identified with the first principle component on a factor analysis of cognitive tasks and, most importantly, should not be regarded as properly expressed by a metric construct, something measurable by a privileged test instrument. A Binet-type test will give a broad classification reflecting some crudely understood aspects of mental development, which still lacks expression in an appropriate concept, but it does not measure anything. (144-146)
This is just like what Howe (1997: 6) states:
A psychological test score is no more than an indication of how well someone has performed at a number of questions that have been chosen for largely practical reasons. Nothing is genuinely being measured.
This prompts Richardson (1998: 127) to conclude:
The most reasonable answer to the question “What is being measured?”, then, is ‘degree of cultural affiliation’: to the culture of test constructors, school teachers and school curricula.
Conclusion
Thus, the fundamental dissimilarity between psychological traits and physical measures has significant implications for so-called psychological measurement in psychology and other social sciences. Physical measures are relatively straightforward due to their objective and quantifiable nature (we can come to similar measurements on a piece of wood for example), while psychological traits are immaterial and and subjective, this means that science can’t study first-personal subjective states.
From the discussion of what a specified measured object, object of measurement and measurement unit are to examples of actual physical measurements that meet these criteria, it is quite clear that IQ—nor any psychological trait—is like a physical measure. While IQ tests are said to be measurement devices, the claim fails upon closer conceptual analysis, since there are no measurement units, and since even before a measurement unit is presented, it must be know whether or not it is possible to measure what one desires to. Psychological traits aren’t actually quantitative since they lack a specified measured object, object of measurement and measurement unit.
If psychometricians have the ability to measure psychological traits using psychological tests, then there must be a specified measured object, object of measurement and measurement unit. There is no specified measured object, object of measurement and measurement unit for any psychological trait. Therefore, psychometricians don’t have the ability to measure psychological traits, and so psychometrics isn’t measurement. Not even the hypothetical construct g (“general intelligence“) will save psychometry. If psychological traits can be measured, then they are similar to physical measures that have a specified measured object, object of measurement and measurement unit. Psychological traits are not similar to physical traits that have a specified measured object, object of measurement and measurement unit. Therfore, psychological traits are immaterial and and so immeasurable.
Nothing is genuinely being measured by IQ test, if we take measurement to be the process of quantitatively determining the value or magnitude of a physical, chemical or other property of a physical object or phenomena, since psychological traits aren’t physical, they are immaterial. And since they are immaterial, then they are immeasurable. Therefore there can’t be a science of the mind. So the claim that IQ tests measure something is false, since there is no specified measured object, object of measurement and measurement unit for IQ. And so, the quest for a scientific foundation for psychology is impossible, most importantly since the mental is irreducible to the physical.
It Is Impossible to Breach Our Mental Privacy Using AI and fMRI
2300 words
Introduction
Recent headlines on AI and so-called mind reading have been extraordinary. “AI can now read minds, Japanese scientists’ experiment sparks ethical debate“, “‘Mind-reading’ AI: Japan study sparks ethical debate“, “Goodbye privacy: AI’s next terrifying advancement is reading your mind“, “Scientists in Texas developed a GPT-like AI system that reads minds“, and “A Brain Scanner Combined with an AI Language Model Can Provide a Glimpse into Your Thoughts” are some titles of recent articles that make this outlandish claim. Claims like this are clearly ridiculous. They assume that through reading neuroimages of our brains, that we can then see what one is thinking. This is hopelessly confused. I will argue here that these claims don’t pass any muster and that’s due to the irreducibility of the mental.
A new article was published yesterday in Nature Neuroscience with the title Semantic reconstruction of continuous language from non-invasive brain recordings (Tang et al, 2023). AI hype has been growing over the past few months due to ChatGPT, and this new undertaking uses AI and fMRI to “read thoughts” through translating brain activity into semantic reconstructions. This is a gross kind of reductionism of mind to physiological brain activity (CNS). But since it’s impossible to localize cognitive processes in the brain, along with the privacy of the mental, then these undertakings are bound to fail. I will argue that it’s impossible for AI to mind-read and that our mental privacy will never be breached.
fMRI and AI
fMRI measures changes in blood flow and oxygenation in different brain regions which allows researchers to see which areas of the brain are more active during the action of cognizing. The assumptions of fMRI to localize cognitive processes, however, fail (Uttal, 2001, 2012, 2014). They fail for a modicum of reasons like individual differences in brain imagine aren’t stable, and so averaging (pooling) disparate studies obscures inter- and intra-subject variation. They are merely reporting random and quasi-random fluctuations in a complex system. Thus, if individual brain physiology is different second to second, minute to minute, hour to hour, how can we logically state that by pooling these images together we can derive where these cognitive processes are occurring in the brain? So the claim that fMRI can localized cognitive processes is false.
It looks like the AI hype train won’t end soon. Like with LAMDA, and ChatGPT, this looks like it will make headlines for a while. But is it true? I will argue that it isn’t, since the mental is private. We have privileged access to our intentional states.
Such articles like this Guardian article, titled AI makes non-invasive mind-reading possible by turning thoughts into text is the newest article reporting on such studies that make these outlandish claim. (In 2018, Mind Matters covered similar AI hype.) The article quite clearly assumes that thoughts are a physical process or a function of physical processes. The fact of the matter is, the paper does not in any way show that AI large language models (LLMs) can read minds. Thoughts are not something that can merely be read based on looking at brain physiology. The Guardian article states:
An AI-based decoder that can translate brain activity into a continuous stream of text has been developed, in a breakthrough that allows a person’s thoughts to be read non-invasively for the first time.
This claim, however, fails and it fails due to a priori considerations. In his paper Immaterial Aspects of Thought, Ronald Ross (1992) (also see Feser, 2013) argued that formal thinking is incompossibly determinate but no physical process or functions of physical processes are incompossibly determinate, so thoughts aren’t a a physical or functional process and no physical process is formal thinking so this then refutes functionalism and physicalism. Here is how Ross (1992: 137) puts it:
Some thinking (iudgment) is determinate in a way no physical process can be. Consequently, such thinking cannot be (wholly) a physical process. If all thinking, all judgment, is determinate in that way, no physical process can be (the whole of) any judgment at all. Furthermore, “functions” among physical states cannot be determinate enough to be such judgments, either. Hence some judgments can be neither wholly physical processes nor wholly functions among physical processes.
This is clearly a form of substance dualism. So thinking and judgment are mental processes which cant be reduced to physical or functional processed and explanations. So this argument has considerations for claims that we can use AI and fMRI to read minds. For if cognition isn’t able to be localized to certain parts of the brain, and if thoughts aren’t a a physical or functional process and, then the endeavor to read minds will.l ultimately fail.
fMRI can, of course, detect brain functioning. However, it can’t detect mental functioning, since the mental is irreducible to the physical (meaning states of the brain and CNS). Mind reading, then, would consist in detecting the content one’s mental states. This, of course, would include one’s subjective states like their beliefs, desires, and intentions. So brain imaging detects brain functioning, but since mind isn’t identical to the brain or its states—that is, since the mental is irreducible to the physical—then such reductive materialism and types of mind-brain identity are bound to fail. (See Glannon, 2017) Philsopher of mind Ed Feser puts it like this in his article Mindreading?:
Might the detection of some other kind of neural pattern amount to “reading” someone’s thoughts? No, for (among other things) the reasons outlined in my series of posts on short arguments for dualism. In particular (as I argued here), given a mechanistic (i.e. final causality-denying) conception of the material world, any material process must be devoid of intentionality. But thoughts are inherently intentional. Hence nothing detectable in any purely material processes (again, where “material” is understood in mechanistic terms) could possibly reveal the content of any thought.
…
This leaves it open that, at least given certain background assumptions, we might guess with some measure of probability what someone is thinking. Indeed, we can do that already, just by observing a person’s behavior and interpreting it in light of what we know about him in particular, his circumstances, human nature in general, and so forth. And of course, further knowledge of the brain might give us even further, and more refined, resources for making inferences of this sort. But what it cannot do even in principle is fix a single determinate interpretation of those thoughts, or reduce them entirely to neural activity. So, no entirely empirical methods could, even in principle, allow us to “read” someone’s thoughts in anything more than the loose and familiar sense in which we can already do so.
These outrageous claims assume that thinking is a physical process or a function of physical processes, when it’s quite simply impossible for them to be. These kinds of studies assume a kind of mind-brain identity, which is falsified by multiple realizability arguments. (It should also be noted that computational models of the mind are also invalid; Tallis and Aleksander, 2008.)
The fact of the matter is, we have private access to the contents of our minds—it is completely internal. Mind privacy is different from brain privacy; of course we can look at the brain’s neurophysiology, but since there is non-identity between mind and brain, this means that it’s impossible to read minds just from looking at brain states (Gilead, 2014). Gilead concludes:
If the mental is irreducible to the physical, brain privacy does not entail mental privacy. Moreover, if the mental is irreducible to the physical, there is certainly more to persons than their bodies.
…
My arguments above clearly show that brain imaging allows no access to our mind and that mind privacy is quite different from brain privacy, as the latter can be breached by brain imaging, whereas the former cannot. We should not worry whether brain imaging can or will be able to read our mind. We have nothing to worry about regarding our mental privacy, for there is no external access to one’s mind. Each of us has exclusive access to his or her own mind. I also show above that a reduction of the mind to the body inescapably fails, as there is a difference of categories between mind and body or brain, which is compatible with their inseparability.
The mental is irreducible to the physical (including, of course, thinking) and science (third-personal) can’t study mind (first-personal subjective states), so these claims outright fail on a priori grounds.
Arguments for mind-privacy
Using fMRI and AI to read minds isn’t possible now, and it won’t ever be possible.
Here is an argument that mind reading itself isn’t possible:
P1: If mind-reading were possible, then people would be able to read others’ thoughts accurately.
P2: People cannot read others’ thoughts accurately.
C: Therefore, mind-reading is impossible.
Premise 1 is the basic definition of mind-reading. It refers to the ability to accurately perceive the thoughts of others. If it were possible, then people would be able to accurately ascertain the thoughts of others. So the accuracy of mind-reading is a necessary condition for it to be possible.
Premise 2: While we can infer what others are thinking based on their behavior, language, and certain other cues, we cannot accurately perceive one’s thoughts since they are not directly accessible. People also have different interpretations of the same cues.
So the Conclusion then follows that mind-reading is impossible. Since the accuracy of mind-reading is a necessary condition for it to be possible, then the lack of the ability makes it impossible.
P1: If it were possible to read minds using AI and fMRI, then we would have clear and consistent evidence of this ability.
P2: We do not have clear and consistent evidence of this ability.
C: Therefore, it’s impossible to read minds using AI and fMRI.
P1: If it were possible to breach one’s subjective mental states, then someone would be able to access another person’s thoughts or mental processes without their consent.
P2: It is not possible for someone to access another person’s thoughts or mental processes without their consent.
C: Therefore, it is impossible to breach one’s subjective mental states.
Mental privacy refers to the right of one to keep their thoughts private, and breaching this privacy would require accessing thoughts in some other way. But it’s not possible to access one’s thoughts on this way, and brain imaging technologies don’t do this since mind isn’t identical to brain.
P1: If mind-reading using AI and fMRI were possible, then there would be consistent and reliable patterns in the brain that correspond to different thoughts.
P2: If there were consistent and reliable patterns in the brain that correspond to different thoughts, then AI algorithm la would be able to accurately interpret them.
P3: There are no consistent patterns in the brain that correspond to different thoughts, since mind-brain identity is false.
C: Therefore, mind-reading using AI and fMRI is impossible.
The irreducibility of mind to brain and the falsity of mind-brain identity theory means that there can be no consistent and reliable brain patterns that correspond to different thoughts.
Case 1: If it were possible to read minds using AI and fMRI, then there would be physical evidence in the brain that corresponds to specific thoughts or mental processes.
Case 2: If it were not possible to detect physical evidence in the brain that corresponds to specific thoughts or mental processes, then it would not be possible to read minds using AI and fMRI.
Case 3: There is no physical evidence in the brain that corresponds to specific thoughts or mental.processes (due to what we know about the multiple realizability of psychological traits).
C: Therefore, it is impossible to read kinds using AI and fMRI (by proof by cases, case 1 and case 3).
There is no empirical evidence to support case 1, and we know that it’s not possible to detect thoughts or mental processes based on brain physiology alone. As case 2, the absence of physical evidence linking brain and mental states 1-to-1 would mean that AI/fMRI cannot detect them. This also suggests that the brain isn’t a purely mechanistic system, which can be fully understood and predicted using computational models. This is similar to Libet experiments, in which it was claimed that unconscious brain activity preceded conscious intention to move; the brain does not initiate freely-willed processes (Radder and Meynen, 2012). Lastly foe the third case, neuroimaging studies consistently fail to detect specific thoughts or mental states from brain states alone. And even if patterns of brain activity can be associated with certain mental states, it’s impossible to determine with certainty what specific thoughts or mental processes a person is experiencing.
Conclusion
While our technology is quickly increasing, a priori arguments show that the explanatory gap between science and subjective mental states is impossible to close. Due to the radically different properties the mental and the physical have, this means that we can’t use science to study our subjective mental states. While there is a ton of fanfare recently about LLMs and the ability of them and fMRI to show that mind-reading is possible, these claims are nothing but hot air. For if the mental were reducible to the physical, then it could be possible in principle that we could read minds based on neurophysiology and brain images. However, since the mental is irreducible, then we can’t use these technologies to read minds.
These claims, though, will increase in frequency since physicalist views are held by the super majority. However, the arguments here show that mind-reading using AI and fMRI is impossible, since mind and brain are not identical.
Thus, our mental privacy is safe from physical systems that attempt–in vain—to breach it.
NotPoliticallyCorrect 