What distinguishes our minds from those of other creatures?

The September issue of the Scientific American (the 'Origins Issue') has an article by Marc Hauser in which he makes a list of what he considers distinctive human competencies (enter 'Hauser' in the search box in the left column to see my previous posts mentioning Hauser's work). Here is a clip from the article:

Although humans share the vast majority of their genes with chimps, studies suggest that small genetic shifts that occurred in the human lineage since it split from the chimp line produced massive differences in computational power. This rearranging, deleting and copying of universal genetic elements created a brain with four special properties. Together these distinctive characteristics, which I have recently identified based on studies conducted in my lab and elsewhere, constitute what I term our humaniqueness. The first such trait is generative computation, the ability to create a virtually limitless variety of “expressions,” be they arrangements of words, sequences of notes, combinations of actions, or strings of mathematical symbols. Generative computation encompasses two types of operation, recursive and combinatorial. Recursion is the repeated use of a rule to create new expressions. Think of the fact that a short phrase can be embedded within another phrase, repeatedly, to create longer, richer descriptions of our thoughts– for example, the simple but poetic expression from Gertrude Stein: “A rose is a rose is a rose.” The combinatorial operation, meanwhile, is the mixing of discrete elements to engender new ideas, which can be expressed as novel words (“Walkman”) or musical forms, among other possibilities.

The second distinguishing characteristic of the human mind is its capacity for the promiscuous combination of ideas. We routinely connect thoughts from different domains of knowledge, allowing our understanding of art, sex, space, causality and friendship to combine. From this mingling, new laws, social relationships and technologies can result, as when we decide that it is forbidden [moral domain] to push someone [motor action domain] intentionally [folk psychology domain] in front of a train [object domain] to save the lives [moral domain] of five [number domain] others.

Third on my list of defining properties is the use of mental symbols. We can spontaneously convert any sensory experience—real or imagined— into a symbol that we can keep to ourselves or express to others through language, art, music or computer code.

Fourth, only humans engage in abstract thought. Unlike animal thoughts, which are largely anchored in sensory and perceptual experiences, many of ours have no clear connection to such events. We alone ponder the likes of unicorns and aliens, nouns and verbs, infinity and God. Although anthropologists disagree about exactly when the modern human mind took shape, it is clear from the archaeological record that a major transformation occurred during a relatively brief period of evolutionary history, starting approximately 800,000 years ago in the Paleolithic era and crescendoing around 45,000 to 50,000 years ago. It is during this period of the Paleolithic, an evolutionary eyeblink, that we see for the first time multipart tools; animal bones punctured with holes to fashion musical instruments; burials with accoutrements suggesting beliefs about aesthetics and the afterlife; richly symbolic cave paintings that capture in exquisite detail events of the past and the perceived future; and control over fire, a technology that combines our folk physics and psychology and allowed our ancestors to prevail over novel environments by creating warmth and cooking foods to make them edible.

Conscious Reasoning and Intuition in Moral Judgment

Cushman, Young, and Hauser. at Harvard, ask...

Is moral judgment accomplished by intuition or conscious reasoning?

They

...investigated three principles that guide moral judgments: (a) Harm caused by action is worse than harm caused by omission, (b) harm intended as the means to a goal is worse than harm foreseen as the side effect of a goal, and (c) harm involving physical contact with the victim is worse than harm involving no physical contact.

They note that

A critical ingredient missing from the current debate is an experimental method that clearly links data on moral judgment with data on moral justification. Without establishing that an individual uses a specific moral principle, it makes little sense to ask whether the content of that principle is directly available to conscious reasoning. Therefore, in the present study, we first identified three moral principles used by subjects in the judgment of moral dilemmas, and then explored the extent to which subjects generated justifications based on these principles....Asking whether these principles are invoked to explain moral judgments, we found that subjects generally appealed to the first and third principles in their justifications, but not to the second.

These experiments support the view:

that moral judgment can be accomplished by multiple systems: Some moral principles are available to conscious reflection—permitting but not guaranteeing a role for conscious reasoning—whereas others are better characterized by an intuitionist model.

Take Marc Hauser's Moral Sense Test

Monkeys prefer silence...

The fact that monkeys prefer silence suggests that humans' music responses may reflect evolutionary selection for cognitive processes linked to emotion and motivation. Here is the brief review from the Random Samples section of the Aug. 3 issue of Science:

Cognitive scientists Joshua McDermott of the Massachusetts Institute of Technology in Cambridge and Marc Hauser of Harvard University put tamarins and marmosets in an apparatus with two chambers, each rigged to play music whenever an animal entered. In one experiment, the musical choices were a flute lullaby (65.26 beats per minute) and Alec Empire's electronic techno hit "Nobody Gets Out Alive" (369.23 beats per minute). The monkeys spent an average of about two-thirds of their time on the lullaby side, showing that they prefer slower tempos. But when given the choice of silence, lullabies, or a Mozart concerto, they spent most of their time avoiding music altogether. A similar experiment with eight humans showed a distinct preference for music--especially lullabies--over silence, the authors report in the September issue of Cognition.

"The observations suggest that only humans have a natural, or innate, inclination to engage with music," says Isabelle Peretz of the University of Montreal in Canada, who has concluded from studies of people with amusia (tone deafness) that humans have special brain pathways for music (Science, 1 June 2001, p. 1636). McDermott and Hauser--who earlier found that monkeys have no preference between harmonious and dissonant music--suggest that humans' music responses may reflect a "unique evolutionary history of selection" for cognitive processes linked to emotion and motivation.

Evilicious humans, hard-wired for war?

"Evilicious" is the title of a new book self published by Marc Hauser, the Harvard Psychologist forced to resign his position after it was discovered that he had been falsifying behavioral data to fit his foregone conclusions. He asks "why seemingly normal people torture, mutilate, and kill others for the fun of it — or for no apparent benefit at all." He suggests that humans uniquely evolved this capacity because "evildoers emerge when unsatisfied desires combine with the denial of reality, enabling individuals to engage in gratuitous cruelty toward innocent victims. This simple recipe is part of human nature, and part of our brain’s uniquely evolved capacity to combine different thoughts and emotions." Hauser's 'trying for a second chance' book has drawn a number of favorable reviews. One summary from Atran's review cited by Dobbs:

....“addiction to evil” – the persistent subjection of innocents to gratuitous cruelty — emerged as a by-product of the human brain’s unique evolutionary design. The ability to creatively combine all manner of thought and emotion enabled our species to produce great works of art and science, as well as to freely choose to kill and torture with a level of maliciousness unprecedented in the history of life on earth. Here we find that the most dangerous and effective evildoers are not sadists or serial killers with disordered minds, but mostly normal people who could have chosen not to kill and torture. When driven by unsatisfied desires — especially if channeled into dreams of glory for a cause — and in denying the reality and the humanity of others, even nice guys can become massively bloodthirsty.

A related issue is whether we are "hard-wired for war". Evolutionary biologist David Barash argues there is no convincing evidence for this. Despite examples such as Yanomamo ferocity, a broad survey shows that peacemaking is, if anything, more pronounced and widely distributed, especially among groups of nomadic foragers who are probably closest in ecological circumstance to our hominin ancestors.

How the mighty have fallen...

I am slack-jawed with amazement on reading that Jonah Lehrer, who I have had great respect for, a brilliant popularizer of psychology and brain neuroscience (“Proust was a Neuroscientist”and "How we Decide") resigned as a staff writer for the New Yorker magazine on Monday after a report that he had fabricated quotes from Bob Dylan in the book “Imagine,” published in March. This reminds me of Harvard animal psychologist Marc Hauser’s fall from grace after the discovery that he had falsified data. How could such intelligent and original people, who write with such clarity and lucidity, be so stupid? I guess ambition trumps all.

Added note: Today's New York Times has an interesting piece on the Lehrer/Dylan affair.

The possibility of impossible cultures.

Marc Hauser offers an essay in the 9 July issue of Nature, in which he suggests points of contact between work in the generative tradition of linguistics (Chomsky, etc.) and evolutionary developmental biology research on animal forms. Just as a developing animal form faces a massive range of possible variation:

...children are born with the capacity to acquire a wide range of possible languages, as opposed to specific languages such as English, Korean or French. This implies that a child is equipped with an abstract acquisition device, allowing the 'growth' of many different languages. Furthermore, as the child's acquisition device generates a space of possible languages, something internal or external to the device creates a space of impossible languages — forms that are never entertained by the child because they are poorly designed for acquisition and externalization in linguistic communication.

...in the same way that biologists speak of morphospaces — n-dimensional volumes that define the range of existing and potential morphological variation — linguists can speak of 'linguaspaces'. These are n-dimensional environments that constrain the set of possible languages and therefore, by definition, establish the set of impossible languages. What is necessary, therefore, is to establish the set of parameters that allow the range of variation and place constraints on its overall form.

...a point of contact concerns how the internal language system ultimately forms an acquired and externalizable language. If, as discussed earlier, the acquisition device constrains the range of possible languages by providing a set of options, then the role of environmental input is to favour, and thus select, certain options over others. This selective perspective, although uncommon in the mind sciences, aligns more closely with other work in biology, including studies of the immune system, the development of animal forms, the wiring of neurons and the acquisition of bird song. For example, songbirds have evolved brains with a set of developmental options for creating variation in song-relevant acoustic forms. Depending on the environment, certain note types are selected and are then reproduced in particular orders to create population-specific dialects — and so it is for language acquisition by humans. When a child is exposed to a particular linguistic environment, the relevant linguistic input or experience fixes the available options to create an externalizable language that is comprehensible to those who will care for and compete with the child.

Research in the generative tradition of linguistics suggests therefore that, like the variety of animal forms, the sense of unbounded variation in linguistic form is illusory, concealing a suite of universally held, biologically instantiated mechanisms for generating variation, allowing acquisition and constraining the space of possible languages. Although biologists have long sensed the close connection between the generative properties of language and generative biological systems, including the immune system, microbial diversity and proteonomics (the study of protein function and expression) relatively few students of the mind sciences have acknowledged such connections with other domains of human knowledge.

Heuser suggests that only humans have evolved four computational capacities, constituting a phylogenetic mind gap between humans and other animals:

Generative computation Recursive and combinatorial operations provide the only known mechanisms for generating an almost limitless variety of meaningful expressions, whether mathematical, linguistic, musical or moral. Recursion is an iterative operation, in which a rule is called up repeatedly to create new expressions, be they embedded phrases within a sentence, new musical scores with repeating themes, or tools within tools (for example, a Swiss army knife). Each expression has a unique interpretation or function depending on the arrangement of the elements. By contrast, combinatorial operations allow discrete elements to be unified and ordered, thus creating new ideas, which could be expressed as novel words (Walkman from walk and man) or novel musical forms.

Mental symbols Humans readily, without instruction, convert sensory experiences and abstract thoughts into externalized symbols, either as words or images. This capacity cuts across domains of knowledge and sensory experience, enabling humans to express beliefs in sentences, to depict particular melodies with explicit notations, and to provide logos indicating when to turn off the highway for a hamburger or a coffee.

Promiscuous interfaces Humans have unique creative capacities and problem-solving abilities, which stem from the capacity to combine representations promiscuously from different domains of knowledge. For instance, humans can combine the concepts of number, belief, causality and harm in deciding that it is sometimes morally obligatory to harm one person to save the lives of many.

Abstract thought Some thoughts derive from direct sensory experiences: for example, thinking of red items such as cherries and blood requires experience with these, as opposed to non-red objects such as celery and bone. But many human thoughts are abstract, with no explicit or even necessary sensory connection. These include concepts such as infinity, grammatical categories such as nouns and verbs, and ethical judgments such as permissible and forbidden.



Figure - Promiscuous interfaces between different domains of knowledge. An example of deciding whether taking one life is justified if the action saves many other lives. A representation of an action (a finger pulling a trigger) interfaces with a representation of death as a potential consequence, which in turn interfaces with a system of numerical representation that evaluates whether the number of lives killed exceeds the number saved. This then interfaces with a moral evaluative system that judges the permissibility of the initial action, which then interfaces with the human linguistic system to deliver the judgement "forbidden".

The generative mechanisms that underpin so much of human mental life acquire their expressive power because the recursive and combinatorial operations can functionally 'grab' the outputs of different modular systems or domains of knowledge. This capacity for promiscuously creating interfaces between domains is almost absent in animals. Thus, although both human and animal brains are characterized by modular functions and mechanisms, the modular outputs are typically restricted to a single functional problem in animals but are broadly accessible in humans. Non-human animals therefore show a form of myopic intelligence, designed to solve one problem with exquisite efficiency. For example, although honeybees have a symbolic dance that indicates the distance, direction and quantity of food, this communication system is largely restricted to food despite the intricate social lives of bees. Although meerkat adults teach their pups how to kill scorpion prey by providing them with age-appropriate opportunities for handling and dismembering, teaching does not occur in any other context. Although plovers use a deceptive display to lure predators away from their nest of eggs, they do not deceive in any other situation. And although chimpanzees use the direction of another's eyes to guide strategic competition, they are far less skilled at using another's eyes to guide cooperation. By contrast, in humans, neither language, teaching, deception, or the use of seeing to infer knowing are restricted to a single context.

Primate beginnings of morality

Nicholas Wade, in the NY Times science section, presents a nice summary of the views of Franz de Waal, Marc Hauser, and other regarding moral behaviors in primates that are antecedent, and similar, to our own. Here is the PDF of the article, and I show the illustrations by Edel Rodriguez based on source material from Frans de Waal and legends below.

Chimpanzees have a sense of social structure and rules of behavior, most of which involve the hierarchy of a group, in which some animals rank higher than others. Social living demands a number of qualities that may be precursors of morality.


Frans de Waal argues that the building blocks of human morality can be seen in the behavior of nonhuman primates like himpanzees. Mutual grooming, for example, shows a sense of fairness or reciprocity.



Chimpanzees engage in both reconciliation and peacemaking. Male chimpanzees routinely reconcile after fights, which protects their group’s social fabric. And Dr. de Waal has described female chimpanzees removing stones from the hands of males about to fight.


The ability to understand the plight of others — empathy — is clearly an essential part of any moral system. Chimpanzees exhibit a variety of behaviors that suggest they have the capacity for empathy, such as helping a frightened young chimp down from a tree.

The origins of morality.

Mark Johnson has generated some creative and seminal ideas in his books "The Body in the Mind: The Bodily Basis of Meaning, Imagination, and Reason" and, with George Lakoff "Metaphors We Live By." I pass on a few clips from Les Beldo's review of his most recent book, "Morality for Humans Ethical Understanding from the Perspective of Cognitive Science":

Over the past 25 years, a growing number of cognitive scientists have taken it as their mission to find an empirical basis within brain science for the distinctive character of moral judgments. Investigators such as Marc Hauser, Steven Pinker, and Jonathan Haidt have posited the existence of an innate, domain-specific moral faculty in humans, be it a “universal moral grammar,” a “moral instinct,” or an “intuitive ethics.”

In Morality for Humans, Mark Johnson introduces an approach he calls “moral naturalism.” It is committed to the idea that moral knowledge does not exist on some separate plane but rather in the everyday habits, practices, institutions, and “bio-regulation” of lived organisms. Johnson is skeptical, however, of claims about the existence of a moral module in the human brain. He notes that “there are simply far too many complexly interacting multifunctional systems … in our intuitive moral judgments for there to be anything remotely resembling a distinct moral faculty.” Although he never uses the term, Johnson argues that moral problem-solving relies entirely on cognitive processes like logic, empathy, or narration. He sees the idea of an innate moral faculty as just another attempt to prove the existence of immutable moral laws, not in divine will or in pure reason but in a strong normative reading of cognitive science and evolutionary biology.

Johnson locates the source of values in our social and biological needs, cultural representations, and personal experiences. Here, Johnson has no qualms about violating the so-called naturalistic fallacy, which suggests that normative statements about how things ought to be cannot be derived from factual statements about what is. He moves freely between descriptions of human needs and human tendencies, on the one hand, and the normative suggestion that we ought to fulfill those needs and support those tendencies. Dismissing the naturalistic fallacy is easy if one thinks of it as an esoteric philosophical concept. But the term refers to a real logical problem of which Johnson is in fact acutely aware: “the fact that we have come to value certain states of affairs,” he writes, “is no guarantee that we should value them in the way we do.” This has been a particular weakness of studies that would make normative claims based on findings in cognitive neuroscience. How can descriptions of how our brains work tell us anything about what we ought to do in particular situations? It is a problem Johnson never resolves.

Morality is typically distinguished from other domains of social judgment by its unconditionality. A moral judgment refers to something that is considered right or wrong in and of itself. Johnson rejects the idea that moral judgments are unconditional, saying instead that the “trumping force” of morality owes to the fact that “certain things tend to matter more for us because they are thought to be necessary for the well-being of ourselves and others.” Individual well-being and societal cohesion are practical ends, however, and concerns about achieving them are matters of prudent conduct and prudent governance. This, along with Johnson's repeated insistence that moral problem-solving is no different in kind from any other form of problem-solving, leads one to wonder why he bothers to retain the concept of “morality” at all. Johnson suggests that values exist only in relation to some predefined or agreed-upon set of goods, feelings, or human needs, but that still creates fertile ground for hypothetical imperatives that are binding upon anyone who accepts the most basic premises of society. Why is this not enough? Is the stigma of moral relativism so frightening? What's so bad about prudence?

A Mea Culpa - Pinker and his critics

I think in general that Steven Pinker goes way overboard on the nativist angle, and so recently approvingly passed on this Churchland review in the Nov. 1 issue of Nature critical of Pinker's new book, "The Language of Thought." - I hadn't actually read the book. These retorts by Marc Hauser and Pinker himself in the Dec. 6 issue make me realize that I should have. I have zapped my original post, and I'm now going to read the book......(one thing about doing a blog is that you read fewer good long books). I admit to a residual grumpyness about Pinker (a brilliant man) from his visit to Wisconsin a number of years ago as a featured speaker. He was dragged through the usual torture of serial 30 minute interviews with local "prominent persons" (I was the Zoology Chair at that time), and during our conversation I found him to be quite remote. At his talk he read from a typescript - word for word - a lecture that I had already heard twice before.

MindBlog retrospective - 3rd and 1st person narrative in personality change

Over the past few weeks, I've been scanning the titles of old MindBlog posts (all 2,588 of them, taken 300 at a time because fatigue sets in very quickly in such an activity), glancing through the contents of the ones I recall as being most interesting to me, and assembling a list of ~80 reflecting some major themes. I am struck by the number of REALLY INTERESTING items that I had COMPLETELY forgotten about. (An illustration is the paragraph starting below with the graphic which is a repeat of a post from May 30, 2007.) It frustrates me that I have lost from recall so much good material. And, of course, it is also frustrating that the insight we do mange to retain will not necessarily change us (the subject of this March 23, 2006 post.)


Benedict Carey writes a piece in the Tuesday NY Times science section (PDF here) reviewing work done by a number of researchers on on how the stories people tell themselves (and others) about themselves do or don't help with making adaptive behavior changes. Third person narratives, in which subjects view themselves from a distance - as actors in their own narrative play - correlate with a higher sense of personal power and ability to make personality changes. First person narratives - in which the subject describes the experience of being immersed in their personal plays - are more likely than third person narratives to correlate with passivity and feeling powerless to effect change. This reminds me of Marc Hauser's distinction of being a moral agent or a moral patient. The third person can be a more metacognitive stance, thinking about oneself in a narrative script while the first person can be a less reflective acting out of the script.

3rd and 1st person narrative in personality change

Benedict Carey writes a piece in the Tuesday NY Times science section (PDF here) reviewing work done by a number of researchers on on how the stories people tell themselves (and others) about themselves do or don't help with making adaptive behavior changes. Third person narratives, in which subjects view themselves from a distance - as actors in their own narrative play - correlate with a higher sense of personal power and ability to make personality changes. First person narratives - in which the subject describes the experience of being immersed in their personal plays - are more likely than third person narratives to correlate with passivity and feeling powerless to effect change. This reminds me of Marc Hauser's distinction of being a moral agent or a moral patient. The third person can be a more metacognitive stance, thinking about oneself in a narrative script while the first person can be a less reflective acting out of the script.

The evolution of morals - a nudge of clarity

Relevant to my recent post, This letter to the editor in the science section of the NYTimes from my Univ. of Wisconsin colleague Larry Shapiro, in the Philosophy Department:

To the Editor:

As a professional philosopher, I found myself with no worries about whether Marc Hauser’s theory would put my colleagues in ethics out of business. There’s a tremendous difference between the two questions: 1) Why do human beings make the judgments of right and wrong that they do?; and 2) Are these judgments correct?

Perhaps an evolutionary story can suffice to answer the first question, but I can’t imagine how it might answer the second.

Larry Shapiro
Madison, Wis.

Recursion in vocalization not unique to humans

Several years ago Marc Hauser, Noam Chomsky and Tecumseh Fitch published an influential paper that speculated that recursion, or self embedding, might be the one aspect of language that is clearly unique to humans. Gentner et al have now shown that the European starling can be trained to recognize complex recursive grammars. A review by Marcus suggest that "the abstract computational capacity of language may consist not so much of a single innovation as a novel evolutionary reconfiguration of many.. ancestral cognitive components, genetically rejigged into a new whole. Contemporary research suggests that the human brain contains few if any unique neuronal types, and few if any genes lack a significant ancestral precedent."

The end of ISMs

Marc D. Hauser:

"Racism, Sexism, Species-ism, Age-ism, Elitism, Fundamentalism, Atheism. These –isms, and others, have fueled hatred, inspired war, justified torture, divided countries, prevented education, increased disparities in wealth, and destroyed civilizations...there is a single underlying cause: a brain that evolved an unconscious capacity to seek differences between self and other, and once identified, seek to demote the other in the service of selfish gains... The good news is that science is uncovering some of the details of this destructive capacity, and may hold the key to an applied solution. My optimism: if we play our cards correctly, we may see the day when our instinctive prejudice toward the other will dissolve, gaining greater respect for differences, expanding our moral circle."

Among the items Hauser suggests in a possible playbook:

"Recognize the universality of our moral intuitions...Remove the doctrinal rules and our intuitive moral psychology propels us to decide what is morally right or wrong based on general principles concerning the welfare of others and our own virtues. If the Protestant and Catholic Irish can see past their religious beliefs, empathize with the other, recognize their shared underlying humanity and settle into peaceful co-existence, why not other warring factions?"

"Be vigilant of disgust...Although disgust was born out of an adaptive response to potential disease vectors, things that are normally inside but are now outside such as vomit, blood, and feces, it is a mischievous emotion, sneaking into other problems, alighting, wreaking havoc on group structure, and then spreading. Throughout the history of warfare, every warring group has tagged their enemy with qualities that are reminiscent of disease, filth, and parasites."