Tag Archives: reasoning

Infants Reasoning About the World

The world may appear to be a ‘blooming, buzzing, confusion’ to infants, but within a few months infants are able to indulge in sophisticated cognition. They develop folk physical/astronomical theory, folk psychological theory,  folk moral theory and folk biological theory, pretty rapidly.

This post is about those cognitive frameworks that infants develop and which more or less persist in adulthood.

An infant

An infant (Photo credit: Wikipedia)

It had been my contention, that Autistic children are predominantly governed by physicalist explanations and frameworks, while those prone to psychosis indulged more in mentalistic cognition.  However, I recently came across the work of Renee Baillargeon and colleagues [pdf, pdf] that suggests that infants have four different types of cognitive frameworks.

  1. Physical reasoning system:  Infants are able to reason about the world like little physicists, assuming objects and their permanence over time.  Thus in a standard Violation of Expectation paradigm, they would look significantly longer at an event where object permanence is violated.  They also assume continuity about the object even when it may be hidden from sight and reappear from behind a blind. Thus they are able to perceive the world in mechanistic terms.
  2. Psychological reasoning system: Infants are also able to reason about the other agents around them, that the agents have a mind of their own and have their wishes, preferences and desires. The infants apply the principle of rationality (consistency and efficiency etc) to the agents and are surprised if the agent behaves irrationally (is inconsistent in his choices or completes an action using inefficient means). The agents are believed to ave autonomous control over themselves and have mental states which make them act as they do. Even at a very young age, infants are able to infer this.
  3. Scoio-moral reasoning system: This sort of reasoning may seem similar to earlier psychological reasoning, but is different in important respects. Infants are able to reason about other individuals that interact with each other. These interacting individuals are governed by principles of reciprocity, loyalty to in-group, fairness etc and infants comprehend and apply these yardstick to interacting individuals and are surprised when these principles are not honored. For eg, they would be more likely to help a helper than a hinderer and would also expect other members to do the same.
  4. Biological reasoning system: This sort of reasoning is about there being certain entities that are a certain form of life viz animals. Animals or life-form is something which has innards or an internal source of energy and if you take the innards out of the animal , the animal stops functioning.  Apart from being self-propelled and agentive, the condition that the animal has innards is very important for this system.

So what does this tell us about humans and the world? I believe this is further evidence that humans have four types of ways of looking out at the world: World in terms of its physical properties and constituents; World as constituting of conscious agents that have metal states that drive their behavior; world as constituting of intentional, interacting individuals that have their inner values and emotions; and finally world as constituting of biological animals that have their carnal grounding. No one way of looking at the world is perfect, The world is both matter and life and we are both minds and morals- it is the multiplicity of the world/ ourselves that makes it so rich and enticing!

Memory and Reasoning: Insight from Apes

I have been reading the excellent book ‘The mind of an ape‘ by David and Ann Premack and also enrolled in a MOOC tiled ‘Origins of the human mind’ offered by Dr. Matsuzawa, so apes have been on top of my mind recently.

Prof Matsuzawa describes an experimental procedure where numerals from 1 to 9 are very briefly displayed on the screen and then masked and the chimpanzee is required to touch the numerals, displayed randomly on the screen briefly, and now invisible as are masked, in ascending order. The chimpanzee is able to perform the task at 80% accuracy, a feat at which if human subjects try they can never succeed (humans perform at 0% accuracy).

We typically pride ourselves as being the epitome of civilization and cognitive abilities, but its humbling to find that there are tasks at which the chimpanzee can excel! This task, in particular, requires immediate memory (sensory/short-term memory) which it seems is better in the chimp.

The different experiments on the chimp also made me think about the underlying structure of memory and reasoning systems. Like humans, it seems chimps too have two different reasoning systems- one tuned to physical world and the other to social/agentic world.

The physical reasoning system is attuned to thinking about causal reasons between psychical objects and events. The question of concern is ‘what caused what?’ . One needs to have a (rudimentary) theory of cause and effect. Some basic understanding of physics is necessary and is instrumental in the development of the capacity of tool use. As a matter of fact too use is one of the ways this physical reasoning system is studied.

The social /agentic reasoning system is attuned to thinking about other con-specifics/ living creatures. It attributes intentions to people and answers ‘who did what to whom?’. One needs to have a (rudimentary) theory of mind to know that others have intentions/ beliefs/ desires etc. A simple paradigm to measure this is whether one understands the visual gaze of a person and can take his/her perspective and know whether the other is able to see something or not.

The physical and social reasoning systems have been show to be different and dissociated in humans and as per one theory are differently accentuated in autistic (more physical reasoning) and schizophrenic (more social reasoning) mind.

Another ability where chimps and humans markedly differ is in their abstract/symbolic representations and linguistic abilities. While chimps can be taught language to a great extent, they don’t develop symbolic language naturally. Language requires abstract and symbolic representation. One can contrast this with the immediate/imaginal representation.

Again, while autistic people have a good immediate/imaginal (thinking/seeing in images instead of words/ symbols) representation system (for e’g’ like in movie ‘rain man’ they can tell the exact number of matchsticks dropped on the floor without counting), their language development is typically hampered , perhaps due to deficits in the abstract/semantic/symbolic representation system.

Thus we see two sets of cognitive functions, and the two sets seem to be slightly at odds with each other: Physical reasoning and concrete/ immediate/imaginal representation; and social reasoning and abstract/semantic/ symbolic representation.

The species  (chimps/humans) who are good at imaginal and physical reasoning system may not be as good at symbolic and the social reasoning system. Similarity within the human family, autistic and schizophrenics may excel at different such functions. While we lost or never gained the ability for highly accurate imaginal system since around 5 MYA when we diverged from chimps and bonobos, we gained the ability for abstract/ symbolic representation. Given the limited real estate that the brain can occupy in any body, its inevitable that as you evolve you lose some and you gain some abilities. Like we lost the ability to use four hands that chimpanzee has.

To summarize, one can associate and link the above to human memory systems. One can conceive of four such memory/reasoning systems:

  1. Visuo-spatial/ short term/ sensory memory: related to immediate memory and imaginal representation.
  2. Procedural memory: related to Physical reasoning/ tool use /physical skills etc and objects representations.
  3. Episodic memory: related to social reasoning and agent representations.
  4. Semantic memory: related to language and symbolism and abstract representations.

Its easy to see how we can apply the same memory/reasoning model to chimps/ other apes without necessarily anthropomorphism. And its equally hard to see and admit that chimps may be better than us at certain cognitive functions and tasks.