AROUND 1970, psychologist Walter Mischel launched a classic experiment. He left a succession of 4-year-olds in a room with a bell and a marshmallow. If they rang the bell, he would come back and they could eat the marshmallow. If, however, they didn’t ring the bell and waited for him to come back on his own, they could then have two marshmallows.

In videos of the experiment, you can see the children squirming, kicking, hiding their eyes — desperately trying to exercise self-control so they can wait and get two marshmallows. Their performance varied widely. Some broke down and rang the bell within a minute. Others lasted 15 minutes.

The children who waited longer went on to get higher SAT scores. They got into better colleges and had, on average, better adult outcomes.

The old adage “good things come to those who wait” isn’t untrue.  Check out this study.

Technorati Tags: Mischel, marshmallow, self-control, success

And by the way, if you get this, you’re a geek. If someone doesn’t do it first, I’m making a teeshirt.

Technorati Tags: psychology, cognition, language, linguistics, avatar, personality, Internet, geek

New analysis of the language and gesture of South America’s indigenous Aymara people indicates they have a concept of time opposite to all the world’s studied cultures — so that the past is ahead of them and the future behind.
[…]

This is a highly interesting phenomenon.  We’ve always taken for an axiom of psychology that peoples all over the world have used the position of “forward” to conceive of the future and “rearward” for the past.  What, then, of these peoples of South America, who defy this axiom?

Technorati Tags: cognition, language, South America, tribe

1.
“It would be an easy matter, of course, to invent a psychological terminology enabling us to describe perceiving, taking, and assuming in sentences which are not intentional. Instead of saying, for example, that a man takes something to be a deer, we could say `His perceptual environment is deer-inclusive.’ But in so doing, we are using technical terms…. And unless we can re-express the deer-sentence once again […] as a nonintentional sentence containing no such technical terms, [the sentence] will conform to our present version of Brentano’s thesis.”

2.
Chisholm is attempting to make a nuanced defense of Franz Brentano’s thesis that the characteristic shared by all mental phenomena, and by no physical phenomena, is intentional inexistence: that when referring to mental acts, we must refer to them as intentional, and not merely in physiological terms. Not only is behavioristic language about reinforcement and physiological processes in the mind too technical for Chisholm, it is also inaccurate because of its deficiency in explaining how perception actually works from a the standpoint of the subject to whom mental phenomena are being presented.

It seems to Chisholm that the only way around using intentional language, especially when describing something about how a person perceives an object in his or her environment, is to use needlessly complex and technical phraseology that does not capture the full meaning of what it is to perceive something. Furthermore, an explanation of perception that does not include intentionality is crippled, according to Chisholm, when explaining how we can take an efficient cause of a presentation to be something that it is not—as in a case, for instance, wherein the man mentioned above could mistake the deer for another animal.

3.
Chisholm begins his article by asking whether Brentano’s intentionality thesis with regard to mental phenomena can also be true of assumptions, and then proceeds to (at least rhetorically) attempt to disprove Brentano’s theory using other peoples’ objections and examples after explaining more fully the terminology Brentano himself was using. For Brentano, as for Chisholm, attitudes and beliefs and other sorts of mental phenomena “intentionally contain an object in themselves,” such that the object presented to consciousness need not exist in real life: I can have a belief about unicorns, or the state of a substance on Twin Earth, or a wish for something that never comes to pass. However, physical (nonpsychological, as Chisholm says) phenomena cannot intentionally contain objects: in order for me to kick a ball, there must necessarily be a ball for me to kick, and so forth.

Chisholm argues that we can talk about states of mind or psychological “directedness” by way of certain types of sentences; in this way he clarifies and re-states Brentano’s original thesis through statements such as, “We may now say that a compound declarative sentence is intentional if and only if one or more of its component sentences is intentional.” Various psychologists and philosophers have tried to re-state the ways of talking about mental phenomena apart from intentionality in various ways, one of which is exemplified by Ayer’s objection that “to think of” something is “to be conscious of the symbols which designate” that thing, but Chisholm says that even this is intentional, since by saying X is designated by Y, we posit nothing about the ontological status or nature of X. Other objections, according to Chisholm, always inevitably refer back to intentional bases, and so assumptions, being mental, must also be intentional.

The overarching point for which Chisholm is attempting to build a case is that in order to describe psychological phenomena, we must use sentences and language that is necessarily intentional, lest we confuse the issues with overly technical language, or by not capturing all there is to a mental act. We can, and according to Chisholm, should, describe physical phenomena from the standpoint of non-intentional sentences; but this is insufficient for psychological language, since intentionality is not reducible to the physical. Therefore, intentional language is the only kind of language adequate for discussing matters of psychology and of the objects of cognition.

Reference: Chisholm, R. M. (1957). “Intentional inexistence.” From Perceiving: A Philosophical Study. New York: Cornell UP.

I’m working in the psycholinguistics lab at the University of South Carolina, so this kind of research is right up my alley. This is a novel study because it shows that people from different dialects of the same language (for instance, Appalachian versus Boston English) have the same kinds of trouble as non-native speakers to a secondary language. (E.g., the Japanese and “L” and “R” sounds in English, rendering the name of this language, “Engrish.”)

This phenomenon also happens in listening to music—if your untrained, Western ears are used to piano and you hear an indigenous Indian instrument (the name of which I have forgotten), there are notes so close together that the untrained ear can’t hear them.

If you’re interested in linguistics, language-learning, or psycholinguistics, check out the article.

read more | digg story

Then the discussion turned to the matter of whether a mathematical language would be able to convey nuance, and whether this was important.

We never got anywhere. Now, I’m odd in that I’m a philosophy major who likes to get somewhere, probably; but this conversation just highlighted for me the whole matter of the way everyone has something to say, and no one listens. It was convicting, really: how often do I pine for a willing ear only to play the part of the overfull mouth? How often do you, gentle reader?

I’ll upload that shortly.

The thesis is, if psycholinguistically we interpret objects of subject-verb pairs more quickly if they fit in context than if not—then do we interpret adverbs in the same way? Hmm.

References:

McKoon, G., Ratcliff, R., & Ward, G. (1994). Testing theories of language processing: An empirical investigation of the on-line lexical decision task. Journal of Experimental Psychology: Learning, Memory, and Cognition, 20, 1219-1228.

Nicol, J. L., Fodor, J. D., & Swinney, D. (1994). Using cross-modal lexical decision tasks to investigate sentence processing. Journal of Experimental Psychology: Learning, Memory, and Cognition, 20, 1229-1238.

Nicol, J. L., & Swinney, D. (1989). The role of structure in coreference assignment during sentence comprehension. Journal of Psycholinguistic Research, 18, 5-20.

Sharkey, A., & Sharkey, N. (1992). Weak contextual constraints in text and word priming. Journal of Memory and Language, 31, 543-572.

I’m not normally one to quote the founding fathers unless it’s early July or a political cocktail party where most people don’t realize what a randy fellow Ben Franklin really was, in his day; but I really liked this quote, especially after what’s been going on lately. This goes along with something that Camus says in his Notebooks about friendship and expectations—but I’ll get there in due time.

I’ve been learning German quickly as possible through the Pimsleur Method. It’s slow but steady going. My professor tried to talk me out of it, though, when he heard that I was starting from scratch with this and I got near-fluency on Spanish lo these many moons ago. I may end up having to do Spanish for my foreign language, but I’d rather just learn that on my own time as need dictates: German is the über-tongue, pun intended.

Incidentally, here is a nice list of the “Best German Films for German-Learners“. Also, an entire series of freely-streamable (but not freely-downloadable) television-based German lessons here. (”Fokus Deutsch” is a classroom method, book not included on the website.)

A parting thought. Is the alienation we feel in America, the sociological WASP problem (i.e., single white male Protestants are the most likely demographic to commit suicide—need I look into Orthodoxy? ), and the general anomie people tend to experience with capitalism—is all of this being addressed by Eastern thought? I have a feeling that what’s pushing most of my peers to dabble in Buddhism (Zen, Theravadan, or otherwise) instead of straight, all-American atheism or some such, is at least the illusion of communion with others, a community atmosphere, a collectivist mindset. Your thoughts?

To that end, it has become fashionable in recent years to use “Seraphim” in relation to oneself as either a screenname, an avatar, or just a “cool thing to which to liken something.” Unfortunately, however cool it sounds, it’s often misused.

Seraphim is the masculine plural form of the Hebrew noun “Seraph” (also transliterated “saraph,” a Hebrew verb meaning “to consume with fire” or, less probably, a noun meaning “a flying, fiery serpent”). Seraphim are the first order of angels, of which traditionally Gabriel and Lucifer are two, and are described in the sixth chapter of Isaiah to have three pairs of wings—one covering their eyes, one their feet, and one by which they are borne up. They are distinct from the cherubim (or “cherubs,” as your or someone else’s grandmother has probably described a gathering of infants), who veil or disclose God, in that they are active, ministering servants.

The Wachowski brothers, whatever else people might accuse them of having screwed up with the Matrix series, got the singularity of Seraph’s name right. Keep it in mind next time you hear reference, and see if it’s understood as plural or singular.

The literal translation, I think, is “abhusten” or “aushusten,” and it means “to come up with (quickly).” Usually it carries violent connotations, or is used in a threat: “Cough up the money by tonight or I’ll break your fingers.” Or, it can be used in a way that implies sympathy toward the actor: “You’re going to have a tough time coughing up more collateral, I know, but I can’t give you the loan without something more substantial.”

I hope that helps!

I updated my BBClone installation today, so the stats were reset and now the Google queries are pouring in. Glad I have readers, even overseas! [Yes, the timestamp is ridiculous. That’s what I get for having too much on my mind and trying to use midday sleep to escape! Ha!]

• Hebrew verb “to kiss” is nashaq, Strong’s number 05401, pronounced with the first “a” like “aw” in “saw,” with the emphasis on the latter half of the word. It’s spelled Nun-Shin-Qoph (also spelled and identical with Nun-Sin-Qoph). Usually used where phileo is used in the Greek NT. See this page for details.
• Setting up a Canoscan scanner under Fedora Core X where X >=2 is fairly easy. Run xsane-config as root, let it yell at you, then try it as a normal user. (This seems to initialize it for me.) According to the Linux USB Scanners page, the Canoscans use a Plustek backend, which you can find here.
• Rafaello Sanzio (Raphael) actually included himself in the far right of the School at Athens (i.e., Scuola di Atene); you can find him beside Sodoma in the far right and center of the painting. He’s looking at you, wearing a cap on his head. (He signified hereby that he was very much on the scientific/Aristotelian end of things.) Here is a closeup of the fellow from this picture. You can find his self-portrait at this link, but the text is in Greek. Poli kalo!

Just in case you didn’t catch all that, I do check the logs (more than I should!), and I do notice who comes to the site and why you’re here. Thank you for your patronage; come back often and let me know if I can do anything to help you.

But don’t stop. Train your eye to go to the next line at least several words in advance of your voice’s arrival there. This isn’t a problem with many poems: for instance, Shakespearean sonnets tend to have punctuation or a certain lilting tone anyway near the ends of lines. But when it comes to some poetry, the effect of flow and continuity of thought can be disrupted severely if you inflect what is not intended by the author.

Secondly, the etymology of the word “barbarian” apparently (according to the Online Etymology Dictionary descends from the sound that the Macedonian peoples’ unintelligible speech made in the Greeks’ ears. I have understood that the origins of the word make the original and literal meaning something along the lines of “wearer of pants,” but while that describes the barbarian peoples (in contrast to the Greeks’ togas, tunics, and general pantslessness), this explanation of the origins of the word makes much more sense—while also providing small but significant insights into how the Greek mind perceived foreign tongues.

Thirdly, the etymology of “matter” descends from Latin materia; that much is clear. But strong connections suggest that materia descends from mater—Latin for “mother.” Now, that puts a certain spin on things, doesn’t it? This is why philosophers try to be very precise with their terminology. Gives a certain spin to “Mother Earth,” doesn’t it? And I must wonder what medieval neoPlatonists did with this concept, what with “God the Father” and Earth, “mater.” Certainly something to keep in mind.

viii.13

Hereagain, we have Augustine expounding childhood development as a byproduct of psycholinguistic training. Interesting to me is how he links memory to intelligence and learning natural language: I have recently been thinking about the link between intelligence and memory myself. Intelligence seems to be largely the ability to recall to mind precisely what fits at that moment in time, whether it be a word or a gesture, a social action or a concept that ties in with another concept. Of course, the difference between “conventional” (however extraordinary) intelligence and autism is, in part, the ability to know when to wield one’s memory, and how. Perhaps, having been away from the texts for too long, I’m stabbing unskillfully in the dark with my own rusty blade of memory.

Natural vocabulary, for Augustine, seems composed of gestures. These are the first symbols we see as infants of the objects we desire, symbols for all objects outside ourselves. Again I think of a discussion on the PCYG in which I argued that, when humanity was in its infancy we required concrete modes of reference (OT sacrifices & ceremonies, Hebrew language lacking abstract words, &c., early development of art, sculpture, music) to that which we felt, thought, understood, and believed. As we grow older, we can understand more and more, what computer programmers call holding more or less “state” in one’s head. In chess this is shown by thinking for the other player and predicting his moves, &c. So what application does this have theologically? What can this tell me about the nature of Christ and of our relationship to Him? Perhaps it is this: that when we are spiritual infants we have to have things concretely spelled out for us—we ask for a briar (it looks pretty, after all) and we see our Father shake His head and offer us laurels instead—but when we are older He sends us cleverly-worded emails that all amount to the same thing: “I’m still in control, beloved, and I am taking care of you.” (That analogy breaks down if you blink, granted, but I’m convinced there’s something there, whether or not I can presently express it.)

Augustine, St. Confessions. Henry Chadwick, trans. ISBN 0-19-283372-3 (Paperback).

This paper is an evaluation of the work done by Arnold, et al. (2004) and Gordon, et al. (1993) and how their work fits together. Some very interesting findings; if I can work out whether it’s legal for me to post the articles, I’ll do so.

• Arnold, J. E., Tanenhaus, M. K., Altmann, R. J., & Fagnano, M. (2004). The old and thee, uh, new: Disfluency and reference resolution. Psychological Science, 15(9), 578-582.
• Gordon, P. C., Grosz, B. J., & Gilliom, L. A. (1993). Pronouns, names, and the centering of attention in discourse. Cognitive Science, 17, 311-347.

Click the title of the post for the PDF version. Be advised, the only revisions I’ve done have been minor reformatting, so if you see the clumsy sentences, be forgiving. Alternately to the PDF, feel free to read more below for the full text. All text that follows is Copyright © 2004-2005 Michael Phillips.

Evaluation of the Effects of Nonstandard Speech on Language Comprehension
Michael C. Phillips
University of South Carolina

Arnold, Tanenhaus, Altmann, & Fagnano, 2004. Abstract.
Most research on the rapid mental processes of online language processing has been limited to the study of idealized, fluent utterances. Yet speakers are often disfluent, for example, saying “thee, uh, candle” instead of “the candle.” By monitoring listeners’ eye movements to objects in a display, we demonstrated that the fluency of an article (”thee uh” vs. “the”) affects how listeners interpret the following noun. With a fluent article, listeners were biased toward an object that had been mentioned previously, but with a disfluent article, they were biased toward an object that had not been mentioned. These biases were apparent as early as lexical information became available, showing that disfluency affects the basic processes of decoding linguistic input.

Gordon, Grosz, & Gilliom, 1993. Abstract.
Examines the ways in which the structure and coherence of discourse are influenced by the manner in which its utterances make reference to common entities. Centering theory, developed within computational linguistics, provides an account of ways in which patterns of interutterance reference can promote the local coherence of discourse. Five reading-time experiments, involving a total of 110 undergraduate Ss, are reported that test predictions of this theory with respect to the conditions under which it is preferable to realize an entity using a pronoun rather than a repeated definite description or name. Data show that there is a single backward-looking center that is preferentially realized as a pronoun, and that the backward-looking center is typically realized as the grammatical subject of the utterance.

[Begin text]

Ever since Noam Chomsky’s groundbreaking work in the 1970s, the field of psychology has been concerned with linguistics: specifically, how we understand and interpret both spoken and written language. A sizable percentage of cognitive research has been dedicated to understanding how both speech and the written word are stored and processed. Until recently, most of this research has focused on how we process and retain fluent discourse; that is, perfectly grammatical language. Many endeavors into the study of language processing have assumed that the most precise and fluent language is always the most informative and the most immediately understandable, but studies have shown that might not be the case. Specifically, the use of anaphora (references in an utterance, or sentence, like pronouns) is a more natural way of referring to some information already given or a subject already presented, as opposed to repeating the exact reference over and over to refer to it (Gordon, Grosz, & Gilliom, 1993); and the use of fluent versus disfluent articles in spoken sentences (for instance, �the�, either drawn out or pronounced quickly) can serve as para-linguistic parameters that actually help interpretation (Arnold, Tanenhaus, Altmann, & Fagnano, 2004).

Regarding disfluency, or �deviations from an ideal delivery� (Arnold et al., 2004), two theories have been advanced by psychologists studying reference resolution in spoken language. One of these holds that any instances of disfluency are hindrances to understanding an utterance; they are, in essence, noise that must be filtered out to arrive at core meaning. The other, advanced by Arnold et al., holds that disfluency is actually a part of the spoken communication process that can be decoded as meaningful information by the listener. Brennan & Schober (2001) as well as Fox Tree (1995, 2001) have found evidence that suggests that bits of an utterance like �uh� and a drawn out, over-pronounced article can signal certain things otherwise left unsaid by the speaker. More to the point, these bits of filler in an utterance, as opposed to the fluent intonation of a phrase, usually convey the kind of information that lets listeners know that some novel input is being formulated or that some novel target is being signaled (Clark & Wasow, 1998; Clark & Fox Tree, 2002; Fox Tree & Clark, 1997). Arnold et al. conducted an experiment to determine how real-time comprehension of spoken language is affected by disfluencies and other non-ideal deviations. They hypothesized that in processing spoken language, listeners immediately use the information provided by disfluency to interpret the meaning of the utterance; and that testing with an eye tracker would show a difference in comprehension and expectation as a function of whether or not the target had been given. Concepts of reference resolution with given information also apply to the written word.

Out of research in the field of artificial intelligence, Grosz et al. (1983) took the terminology and some of the essential framework from the work of Joshi and Weinstein (1981) to advance a theory called centering. Findings from the field of artificial intelligence research supported the idea that, on a global scale of discourse, exact references are most preferable to refer to things, people, or events; whereas, on a local scale, pronouns would usually be generated. This is the crux of centering: the understanding of how anaphoric reference affects the understanding and processing of a local discourse. Broadly, centering theory is based on the idea that there is a �forward-looking center� and a �backward-looking center� in every utterance (except for the first one) in a discourse. These centers of an utterance facilitate the coherence of a local discourse by linking one utterance to the next in a chain of meaning. Two rules govern the theory of centering: the first deals with the ranking of each center in an utterance, and the second deals with how centers are transferred locally across utterances. Gordon et al. (1993) were interested in how centering theory could explain the effects of violation or utilization of the rules of the theory on coherence of a discourse. It was their hypothesis that coherence, as a function of self-paced reading time, would increase across a local discourse with centers (backward- and forward-looking) that pointed to one another, as opposed to the centers being redefined explicitly in each utterance. That is to say, they thought a reference made with a pronoun (e.g., �she�) could provide greater coherence in a discourse than a repeated exact description (e.g., �Susan�).

Method: Study One

A total of twenty-four individuals from the University of Rochester, either as students or faculty, participated in the study. All were native English speakers. Further demographic information was not given. Method of sampling is unknown, but was likely volunteer-based: each participant was paid $7.50 for being a part of the study.

Participants wore an eye tracker while watching a computer screen and were given instructions to use a mouse to place objects in certain positions that they would hear; the experimental variable being tracked was the position of the eyes during (and after) participants heard the prerecorded utterance. The screen would hold two items whose beginnings were phonetically similar, like �candle� and �camel,� called cohort competitors; the other two pictured items were distractors phonetically unsimilar to the others. Participants heard two sentences: the first introduced the target, and established it as �given�; and the cohort competitor would be referenced in the second sentence. These two configurations formed the basis of four conditions: the second sentence could be a fluent or disfluent utterance with reference to either a given or new target. Eight lists, randomly ordered, were comprised of sixteen experimental and thirty-two filler utterances. The independent variables were disfluency, the state of the target as given or new, targets themselves, and item groups.

Data from the eye tracker shows that �looks� to each on-screen object under each condition were strongly linked to whether or not the reference had been fluent or disfluent. Analyzing the data about the looks from 200ms to 600ms after the start of the target noun in the utterance, Arnold et al. (2004) found statistically significant differences in each experimental condition. When the utterance was fluent, participants showed a bias toward a given target, such that when a new target was mentioned fluently, they had already begun looking at the competitor (which had been given). Likewise, with the disfluent utterances, new targets were expected and the competitor was again preferred when expectations were violated; i.e., when the target had already been given. Essentially, articles were shown to be processed on-line to facilitate understanding of utterances.

Method: Study Two

Gordon et al. (1993) conducted five experiments for their article. All participants were students at Harvard University who responded to posted notices for single sessions, each paid. No students were used twice. Twelve students were used in the first experiment; sixteen in the first half and eighteen in the latter half of the second experiment; sixteen in the third experiment; twenty-four in the fourth experiment; and twenty-four in the fifth experiment.

All experiments used sentences displayed on a computer screen, with each broken up into �subjectively natural boundaries� (Gordon et al., 1993), be they phrases or whole sentences. Participants were instructed to read each sentence at a natural pace and tap the space bar whenever they were done reading each section of a sentence, after which a comprehension test was given to control for variance due to ignorance or misinformation on the part of the participant.

The first experiment tested whether centering theory was correct in positing that a backward-looking center in a non-initial utterance would be read faster as a pronoun or an exact description (specifically in this case, a name). The independent variable was whether pronouns were used to refer back to previously-mentioned subjects, making three experimental conditions: pro-pro, pro-name, and name-name. Gordon et al. (1993) performed analyses of variance on the resulting reading times. The pro-pro and pro-name conditions were relatively similar in reading time, but there was a statistically significant slow-down for reading times with the name-name condition, where no pronouns were used. This finding supports their hypothesis.

Further work was done with pronoun placement in experiment two. Five sentences were given to each participant, each in a natural pattern of pronoun usage; the fourth sentence was the independent variable, where either a name or a pronoun was given to refer to a previous subject. There were two parts to this experiment, one using sentences in a block of four, and the second part using sentences in blocks of three. Remarkably, statistical analysis showed no significant gain or loss in reading time for the pronominalized versus name-name conditions. These results contrast sharply with the finding in the first experiment.

Experiment three was conducted so as to find some common ground between the first and second experiments. Passages of three sentences, in which the first sentence used names and the second sentence used a combination of either pronouns or names, or both, were given to each participant. Forty experimental passages were constructed and interspersed with filler sentences; one block of sixteen filler sentences and then �five test blocks of 24 trials each� (Gordon et al., 1993) were given to test these conditions. In this experiment, there was a statistically significant indication that using a name in the test condition instead of a pronoun increased reading (and therefore presumably comprehension) time, which matched more closely the findings of the first experiment.

Fourthly, Gordon et al. (1993) tested centering theory’s handling ideas of continue and shift, that is, whether there is relative continuity or discontinuity across two utterances, as a function of reading time. Additionally, they tested how coherent a backward-looking center of a given utterance would be, depending on the rank of the forward-looking center in the immediately preceding utterance. Passages in this experiment consisted of an initial pair, in which essentially names were introduced; then either a continue or a shift condition; and alternatively another sentence pair was given as a control that did not involve either continue or shift. Analysis showed that using names in a continue condition (where centering theory says is unnatural) increased passage comprehension times, and vice-versa; and that using a pronoun in the shift condition also increased reading times, and vice-versa.

Finally, the fifth experiment attempted to separate the effects of the so-called repeated-name penalty to the backward-looking centers of an utterance and the actual ranking of the previous utterance’s forward-looking center. The design was similar to the fourth experiment, in that it utilized passages of sentence pairs involving either continue, shift, or an alternate construction after the initial pair. Essentially, the results after statistical analysis suggested that surface position of a pronoun would determine its ranking and, thus, its comprehensibility. These results are consistent with centering theory.

Discussion

Both of these articles point out significant issues in cognition and linguistics, but both could also be a bit stronger in their design. For one, though cost can become prohibitive, a larger sample size for each experiment, in the design frameworks of both Arnold et al. (2004) and Gordon et al. (1993). Arnold et al. could have used more data with which to test: more pictures on-screen with more cohort competitors (if not simultaneously on-screen, then more trials per participant using unique icons in each trial). Gordon et al. could also have benefited from more test subjects, but their main problem may have been in their data set. Using sentences of varied complexity and length, they could have tested for how centering theory was affected by utterance length. Also, their terminology was somewhat vaguely defined in certain areas, and using �subjective� sentence breaks might be questionable.

Nevertheless, these articles stand as intriguing inroads to further research. Particularly, work done by Arnold et al. (2004) raises questions about what is and is not �noise� in verbal communication; and Gordon et al. (1993) have done outstanding work that informs both computer science as well as linguistics, especially insofar as when we use pronouns and other designations. †

References

• Arnold, J. E., Tanenhaus, M. K., Altmann, R. J., & Fagnano, M. (2004). The old and thee, uh, new: Disfluency and reference resolution. Psychological Science, 15(9), 578-582.
• Gordon, P. C., Grosz, B. J., & Gilliom, L. A. (1993). Pronouns, names, and the centering of attention in discourse. Cognitive Science, 17, 311-347.

• Brennan, S.E., & Schober, M.E. (2001). How listeners compensate for disfluencies in spontaneous speech. Journal of Memory and Language, 44, 274-296.
• Clark, H.H., & Fox Tree, J.E. (2002). Using uh and um in spontaneous speaking. Cognition, 84, 73-111.
• Clark, H.H., & Wasow, T. (1998). Repeating words in spontaneous speech. Cognitive Psychology, 37, 201-242.
• Fox Tree, J.E. (1995). The effects of false starts and repetitions on the processing of subsequent words in spontaneous speech. Journal of Memory and Language, 34, 709-738.
• Fox Tree, J.E. (2001). Listeners’ uses of um and uh in speech comprehension. Memory & Cognition, 29, 320-326.
• Fox Tree, J.E., & Clark, H.H. (1997). Pronouncing “the” as “thee” to signal problems in speaking. Cognition, 62, 151-167.
• Grosz, B.J., Joshi, A.K., & Weinstein, S. (1983). Providing a unified account of definite noun phrases in discourse. Proceedings of the Fifth International Joint Conference on Artificial Intelligence, Cambridge, MA.
• Joshi, A., & Weinstein, S. (1981). Control of inference: Role of some aspects of discourse-structured centering. Proceedings of the International Joint Conference on Artificial Intelligence, 385-387.

Subject Heading (con’t)

This shows a lack of understanding about the nature of the apostrophe and its function in a word. (Notice, as a related aside, that there’s no apostrophe in “its” here because that would signify a contraction of the being phrase, “it is” and not a neutered possessive singular pronoun.) Of course we all know that what my professor meant was to abbreviate the word “continued.” How would you do that? Like so:

cont’d

That’s because the apostrophe serves to signify content that has been omitted from a word, a phrase, or other language structure. If we take my professor’s example, we have nothing omitted where it is signified, and if we were to take this abbreviation literally, since it does not use a period (”cont.”), we would be left with a word that ends in `t’ and starts with “con”–with some indeterminate number of letters therebetwixt! Observe:

con[?]t.

Now note the structure of the logical reduction signified by the apostrophe in “cont’d”:

continued.

And, have mercy, this same concept applies to that aberrant member of the English language resting on so many Southern tongues like a plague, “y’all.” This contraction is almost invariably written as “ya’ll”. If you’re going to use this abbreviated version of “you all,” at least omit where omission is necessary. The reason for this being misspelled is most likely because “ya” is popularly being used as the auxiliary/slang second-person pronoun of indeterminate number (”you” v. “you”–either could be singular or plural). “Ya” is not a word in English, unless it’s a transliteration of the German affirmative (ja). So here note the examples:

ya’ll becomes ya all

Versus the following:

y’all becomes you all

Thank you, that is all.

Aoccdrnig to a rscheearch at Cmabrigde Uinervtisy, it deosn’t mttaer in waht oredr the ltteers in a wrod are, the olny iprmoatnt tihng is taht the frist and lsat ltteer be in the rghit pclae. The rset can be a taotl mses and you can sitll raed it wouthit porbelm. Tihs is bcuseae the huamn mnid deos not raed ervey lteter by istlef, but the wrod as a wlohe.

Amzanig huh?

Many of you will be familiar with this, psychology majors or not; it made its rounds September 2003 in peoples’ inboxes. This showed up a little while ago, once again, on the discussion boards for USC’s PSYC405 class (Cognition), and I typed up a little response that I thought was relevant. Part of that post follows.

First, the fun stuff:

• A Perl script by Jamie Zawinski that will actually scramble your text for you in this way can be found at the following URL: http://www.jwz.org/hacks/scrmable.pl

  (Discussing this with some comp.sci guys on IRC, all you have to do is leave the first and last letters intact and scramble the internals at random.)

• Aardvark Business (.net) hosts a web-based tool that will scramble your speech in a way that makes it much harder to read, for whatever reason. One head researcher decided to take a contradictory opinion to the one purportedly from the Cambridge guys, saying that if your words are long enough and the letter distribution is intentionally confounding, even if the criteria that the first and last letter be in place and the word be spelled (when decoded) correctly, then you won’t be able to figure it out without a great deal of effort. To prove his point, he got some CGI-savvy fellow researchers to design a tool that does just that. Check it out: http://www.aardvarkbusiness.net/tool/

Matt Davis, who I guess is actually a professor or at least a researcher at Cambridge, wrote up one of those really long, drawn-out, exhaustively comprehensive articles on this like all of those that intellectuals are prone to do when they see something that touches their field. It’s actually pretty fun to read, like getting it straight from the horse’s mouth. It’s pretty long, though, so here are the high spots:

• He postulates that we can see this phenomenon in action in English, especially, because there is so much redundancy in the language (what with vowels & such) and we can get much of the general meaning of the sentence from context. In other languages—particularly semitic ones, e.g. Hebrew—it’s nearly impossible.
• Dr. Davis points to Graham Rawlinson’s doctoral thesis from 1976, “The Significance of Letter Position in Word Recognition”. In a note from Dr. Rawlinson to Dr. Davis, he summarized his points in a very readable, short-page ditty. Cool if you’re thinking about research in that area.