Archive for May, 2003

030515 – Perceptual(?) oddities

Thursday, May 15th, 2003

030515 – Perceptual(?) oddities

In 1969, I remarked to Jerry Fodor, who was then my research advisor at M.I.T., that I had recently had an experience in which I glanced at a scene out the window of a bus, glanced away, and then realized that I had seen some word that seemed unlikely to have appeared in the scene.  I had no idea where the word had appeared, but I felt confident that I had indeed seen it.  I returned my gaze to the scene, scanning it as systematically as I could, looking for the word.  After perhaps five or more seconds, I finally located it.  It was on a sign on a building with many signs.

What struck me as notable was that I knew I had seen the word, but I had no idea where in the scene, I had no idea whether the word was written in large or small letters, or what color the letters were.  All I had was the experience of having seen the word, but beyond knowing what the word was, apparently nothing.  Fodor’s comment to me was that clearly the brain’s system for reading does not mark words it reads with their location in space.  (Although note that students often remember where a certain piece of information was located on a page.)

Since then, I have noticed the same phenomenon, reading a word and not knowing where the word appeared in the visual field without having to mount a conscious search.  Is it right to call this perception without awareness or is it actually awareness without perception.

This morning, while listening to the news on the radio, the phone chirruped briefly as if someone had dialed an incorrect number and almost immediately realized it and hung up.  I noticed that I had no idea whether the sound of the phone had occurred before during or after whatever was being said on the radio at the time, that is, my uncertainty of the temporal relationship between the sound of the phone and the sound of the voice on the radio was uncertain over a range of two seconds or more.  This does not worry me.  I can’t think of a reason I would need to be able to make a finer distinction (except in a psycho-acoustic experimental paradigm).

There is, of course, a line of psycholinguistic experiments that explore variants of this phenomenon.  The subject listens to a sentence and at some point in the sentence there is a beep or a click.  The subject is then asked to identify where in the sentence the sound occurred.  The perceived temporal position of the sound can be manipulated by systematically varying the grammatical structure of the sentence.

030502 – Lakoff, metaphor and metonymy

Friday, May 2nd, 2003

We observe displacement activities in animals.  When an animal is frustrated in an attempt to attain a goal, it sometimes exhibits activity characteristic of an attempt to reach a different goal.  We characterize this as displacement.  Such behavior is sufficiently widespread to conclude that it is at least not detrimental to survival and it may be that an argument may be made for it being favorable to survival.  Arguably, the majority of human time is spent in such displacement activities.

If we grant a certain universality of displacement, at least among animals we regard as closely related to us on the evolutionary scale (e.g., mammals), can we identify patterns of displacement that are analogous to what Lakoff describes as metaphor and metonymy?  In other words, is the human capacity (facility, propensity) to use these patterns (processes) evident in other animals about which it may be easier to think?  Metaphor is, after all, a redirection of attention from one pattern to another pattern which is in some significant way different from and in some significant way similar to the original pattern.

Lakoff insists on a distinction between metaphor, where one thing stands for another, and metonymy, where a part of one thing stands for the whole.  I guess I see metonymy as a special case of metaphor, and I would change its definition slightly to say that in metonymy a part of one thing is used as a metaphor for the whole.

I guess I think in terms of activation of patterns, where a pattern is like a Prolog [a programming language] predicate: it has slots that may or may not be bound to particular instances of things which may themselves be things or patterns.

But more than that, I see humans as pattern-extraction machines.   It may be that the way we understand our environment is that our perceptions and experiences activate patterns that we know about (that we know how to deal with).  After all, that’s pretty much what the visual system does.  It finds patterns and provides gestalts.  A gestalt is a percept.  It comes to us as a whole (a gestalt), but we can inspect its components.

Indeed, gestalt is the prototypical part / whole pattern.

Short-cuts are an essential part of behavior.  There seems to be a race for the effectors or a race to create an effector plan.  First one to show up with a complete plan or plan fragment wins.  Example: I can write the word ‘of’ using the ‘o’ and the ‘f’ I was taught to use when learning to write in script; but I have also acquired a digraph ‘of’ glyph that I can write faster.  Most of the time, I write the digraph for the word ‘of’, but use the two individual glyphs when the letter ‘o’ is followed by ‘f’ in any other context.  Similarly, I have variant forms for ‘s’, ‘t’, and ‘r’.  Whichever I produce, I rarely, if ever, end up creating a monster glyph that is some amalgam or of two forms or a doubling in two forms (which one might expect to happen if multiple motor plans ran at once).

This was part of Lashley’s serial order in behavior paper, although he did not, to my recollection, characterize the process as a race, but that was the gist of it.  Serving as the staging area for serial behavior plans may be one of the functions of short-term memory that David Marr was alluding to when he said, “I expect that there are several ‘intellectual reflexes’ that operate on items held there about which nothing is yet known and which will eventually be held to be the crucial things about short-term memory.”  (Marr 1982, p.348)

Preconscious processing of ambiguous words and phrases slows down when two alternative interpretations are more balanced in likelihood and the ambiguity is more difficult to resolve.  (Paraphrased from Baars 1994, which cites MacKay 1966 “To End Ambiguous Sentences” as the source.  See also Paula M. Niedenthal 2007.  “”Embodying Emotion.”  Science, 316: 1002-1005)

(Lakoff 1987, p452) “Any adequate psychological account of the learning of, and memory for, the human lexicon will have to take account of the phenomenon of folk etymology—that is, it will have to include an account of why expressions with motivating links are easier to learn and remember than random pairings.”

‘”Much in language is a matter of degree”. This section states most strongly Langacker’s conviction that most of the psychological grounding of language uses mechanisms which work by approximation rather than by any type of formal logic. He believes that the basic nature of categorization is well stated by the prototype model and that most distinctions are based on gradients rather than dichotomies. (from review retrieved 030502 of Langacker 1987 Foundations of Cognitive Grammar, Vol. I)