LDT experiments have dichotomous responses because the subject has or does not have a word in his own mental lexicon, so is difficult to make a lexicality graduation. The closest thing to a gradual notion of lexicality can be obtained by studying the lexical frequency of words in terms of velocity. The more frequent a word is, the easier and faster it is accessed in the mental lexicon and vice versa. In fact, a word may be so rare that a subject not even have in their vocabulary, answering in a DL task as a non word. Furthermore, in transparent languages (spanish, for example) the syllable unit can be used to manipulate the frequency syllabic, obtaining much more information about whether a word is more or less rapidly responds in a LDT task like this.

Categorical logic address both syntax and semantics by category, and an interpretation by a functor; I prefer the probability distribution of a gradual notion through the interpretation of context and extra paradigmatic resources.

Actually, not all words are present in our mental lexicon (and does it contain lexical units at all, or maybe it contains only lexical constituents, as in (Liu, Perfetti 2005)?). For example, names are considered words, but at least half of your names I see for the first time (as you mine, I guess). New terms, very old or rare words may seem non-words in isolation, but will become words for us within a context.

And experimental non-words - they are usually made of real words by replacing one or two letters. But one or two letter difference between words is referred to as 'neighborhood' (game - same - name...), the notion that was at first also categorical in psycholinguistics, but then was proved to be a gradual one.

The good question is how to measure lexicality. Probably, like neighborhood - by frequencies of all neighboring words. It would be interesting to have some details from Osmar about probability distribution "through the interpretation of context and extra paradigmatic resources".

Neighborhood is relative to syllable sounds net surroundings, there is always a nearby frequency in terms of sounds and the mind tends to get the maximum close appearance with previous existent , not in reference to meaning but to sounds

( eletrobiochemical frequency scale ).

Thanks for the answers. I think, "true/false" is an artificial approach, not inherent in human nature. That's why Eleanor Rosch had to propose a Prototype Theory instead of strictly categorized way of thinking. And mental lexicon is hardly a stock of ready words, it's rather a stock of word components (constituents), which we easily assemble to make a new word. Most terms were new words for us at their time, including some funny examples like "Monicagate" or "Obamamania", and we apparently didn't have them in our lexicon, but easily recognized them as words. So it is not a simple matter of "true" or "false".

I held some experiments on reading, where the subjects had to recognize letters from different elements of a blurred text (letters, letter strings and the text itself). The longer was the element, the better was recognition of each letter. Of course, in words and in the text there was some facilitation from semantics (word superiority effect, context). But the same was in a sublexical level as well. The effect was especially strong for four-letter strings which were parts of real words but not words: in English they would be like "expe", perc", "reco", etc. Say, letter "e" in these examples would be recognized better than in isolation, in two or three-letter strings. In five-letter strings ("exper", "perce", "recog'...) it wold be recognized a bit more better. I called it a "quadrigram effect", and I think, it is a part of a word superiority effect, an illustration of how it is created. The letters are recognized better because these letter strings are "almost words", that means gradual lexicality.

I like Osmar's comment. Reading in alphabetic languages goes through the circle: "orthography" to "phonology" to "semantics" and again to "orthography" and further at the next level (for Chinese the direction is backward). So neighborhood is really related to sound first, and meaning is the next "station".

Mind works dynamically as a radar no matter whatever may be the stimulus, visual, audible, It(mind) has a direct and objective main goal: make sense according to the capacity of the brain net ( of course each person has a graduated level due to what we call intelectum, the amount of brain memory; cataloged experiences mind -body interconnected) .We can read it into the so called 1895 Freud´s Project. If we take this we will come to the notion that all data received by the mind are perceptible due to one single way of inner transmission. What takes us to Boson. Non words or any kind of inform the mind receives is stored as rests that will keep its own power lasting until nothing may come ahead or will be used as trash material for dreams. But speaking of sounds we could understand why music of any other language than ours within lyrics are kept and one may even sing it without any knowledge of its words ( which are non words of course, but that will be kept as a discernible resource to be used when the mood memory comes back again). Meaning is a parallel process.If we make another parallel, just write down a wrong word into your computer and use the lexical correction of the software. It is not the same, but it is a simile.

As to "through the interpretation of context and extra paradigmatic resources". It is the refined intellectum that start up looking for residues of the environment that may help to make sense for signifying "non words" into possible data of the paradigmatic extra data such as: a smile, a face that does not seem that is the right one for the word;anything that at that same moment may deduce a new sign. By all resources example: if someone say "balls" and keep a hand into a body place, it changes or not the meaning?

Just as if any stranger into a unknown language holds his testicles and say a brief word one can deduce that through extra paradigmatic resources, that he is saying balls...What may take to a "wrong " deduction, but that is the way mind works, trying to make meaning and reestablishing equilibrium.

Once all occidental languages have some common traces and roots, prefix and suffix axiom is very good to be studied under such operational underground of looking for the secondary meaning resource under which they are kept into minds, etymologically speaking we may have a very good idea of the use mind makes of such operational net. It means: a separated group of prefix and suffix storage that works as resources to express some common situations.

Monicagate for example, into Portuguese, due to phonetics takes to moni- ca ( plus gate) induces the non speaker and low informed people to: I sheet for monica, from the Latim- cacare. cacas pro monica, in Russian= дерьмо для Моники. In Turkish kaka means the same..Has the same sounds for the same significant and signified.

Starting from the end: in Russian "kaka" (with first syllable stress) is a baby-talk word for the same process and product, it may be not registered in dictionaries. Football commentators at first avoided to name Brazilian player Kaka for this reason, trying to call him "player #23", 'this Brazilian striker", etc. But then he became very famous, and got his name in Russian with the second syllable stress. Just to make some difference... :)

I thoroughly agree with these words: "Mind works dynamically as a radar no matter whatever may be the stimulus, visual, audible, It (mind) has a direct and objective main goal: make sense according to the capacity of the brain net." One Russian famous linguist (V.B. Kasevich) calls this "presumption of meaningfulness".

I understand now what extra paradigmatic resources and context interpretation were meant. I agree with it. But it's getting far and far from linguistics and even psycholinguistics to psychology and philosophy. At least it is rather far from reading, my field of interest. I appreciate this discussion, it helps to rise over the problem and look wider. Thanks.

Able is a suffix that has its own and specific meaning .

Dart is a new class-based programming language for creating structured web applications- SO, so dartable is not a "pseudo" word ", once your approach is wrong; you have to understand that mind adore new meanings- it is a challenge. It always approaches to and the live language is not under control of any science, it is a mind resource to show up something into a manner that it is not disposable. For live language ( parole ) grammatics does not matter.

Hi, Agnes! I don't quite agree with you in my turn. A person knows a lot of words of his/her native language, but exactly not all. And he/she knows about it, being always ready to encounter a new word. I regularly hear new (for me) words of my native language on TV news, read them in fiction and especially in historic and science literature. For example, medical description for some medicine is full of words that mean nothing or little for me. And I'm not surprised. So subjects in LDT could suppose that some pseudowords are unknown words (if they look like words as 'dartable'). The subjects may try not to look ill-educated and give wrong decision for this reason. But the matter is another: that word superiority effect has gradual nature. It is based on frequency of use and connection to meaning. Sublexical level also has a lot of frequent letter strings, most of which are connected to meaning (morphemes). Is "appro" a word? Of course, no. It is not even a morpheme. But this letter string is quite frequent and associated with only a couple of meanings (approach, approximately, maybe one-two more). Hold an experiment, as I did, and it should show that letters in this string are recognized better than in isolation or in a shorter string (or in infrequent strings without associations to real words). So, the word superiority effect works at sublexical level, that means graduality of the effect itself and of the property it reflects, i.e. lexicality.

I see we are reaching a consensus, at least about LD being a very complex process with different factors. And I think there is a sort of simplification in interpreting results of LD experiments, one of the reasons, in my opinion, is a simplification of the notion of word. If you are a linguist, you know that there is no complete definition of a word. We've been talking about pseudowords, but we write the term differently: you do this as two words, I do as one (because I saw it this way). Is it one word or two? There are a lot of such double-single words in English. In Chinese almost every two characters in a sentence could be considered as one word or two. So I think a word is some degree of integration of orthographic, phonological and semantic information. It is often "more or less" word, because orthographic word may not match with phonetic word and be only part of meaning or conjunction of several meanings. It is a continuum, but it is discrete, at least by letters in orthography. I'd better draw a scheme, but don't know how to do it here. I see it as three interacting hierarchies of the above-named word constituents, with a word as activation reaching some level of the three.

Langue et parole...Langue = categorial notion Imposed

Parole= gradual notion

I'll take an example of "appro" again. No doubt, subjects would answer "no" to this letter string in LDT. But the aim of an experiment is not to check the subjects' knowledge of words, is it? The aim is to measure some variables in two conditions: under influence of lexicality and without it, right? 1 or 0, 100% or 0%. But letter strings like "appro" do have some degree of lexicality, maybe 10%, maybe 50% or even 80% for some people (I don't know how to calculate it correctly, but it is another question). This "hidden" lexicality affects the results of experiments and should be accounted for.

Again, what is lexicality? Frequency of use and connection to meaning. In this regard pseudoword "mable" has more lexicality than the name of one shortly famous Icelandic volcano (Eyjafjallajokull), and letters in the former would be recognized much better than in the latter. Because the pseudoword has a frequent letter string -able, being a morpheme, while the volcano name is not a typical word. I know, no one would use this volcano name in LDT, for this very reason - it is not a typical word (but it is a word). For this reason we use in LDT only nouns - they are "more typical" words than, for example, participles or prepositions.Typical (prototypical) or peripheral are terms for gradual notions, as lexicality, I believe, is.

" but due to status as subparts of actually existing words to which they give activation" this means graduallity , just read sub-parts as letter strings of words already into memory! This is called memory proximity.. This is a mirroring process . The more someone reads, more accuracy and velocity. Lexicaliity is competence...We can arrange a text that only a few letters are exposed of each word and so the paragraph will be understood.I disagree of such approach once it is only a twist of the gradualism of status ( syntax).

Hi, Sebastian! Your explanation about "ABC" is quite in line with how I see it, but conclusions are different. As far as I understand, you take semantic representation (knowledge of a word meaning, lexeme) for the whole word in the mental lexicon, and orthographic representation as something connected to a word, but not constituting it. If so, you should treat phonetic representation the same way. But is it a word that has only meaning without sound? What about the signified and the signifier? (Orthography is a signifier for signifier, a signifier in square, as L.S.Vygotsky called it a century ago). The matter is that any word consists of both sides: meaning and sound, and the written word has the third part: image. They all constitute the word. So if one of the constituents (one of the signifiers) is not complete, but still leads to the signified, is it a word? Does it create a word? I think, yes. Is it a 100% word? I'd say no. It creates a word with high (but not full) probability. For example, "My pen is on the tabl..." There is very high probability to create a word "table" by adding "e", but there's also some probability to create other words like "tablet", "tablecloth", "tableau"... So, letter string "tabl..." has several links to meanings with different probabilities. This link between the signifier and the signified I believe to be lexicality - an ability of being a word. The more links, the less lexicality (ta..., tab..., tabl..., table... have different amount of possible endings). The only link means full lexicality, the exact word. This is the way I see it.

It is not absolutely monolithic, but surely not leveled like in these models. They are a bit old. I prefer connectionist approach, as of M. Seidenberg. He has a lot of publications (http://lcnl.wisc.edu/publications/?member=marks). I'd recommend to read: Seidenberg, M. S. (in press). Computational models of reading: Connectionist and dual-route approaches. In M. Spivey., K. McRae., & M. Joanisse. (Eds.), Cambridge Handbook of Psycholinguistics. Cambridge University Press (http://lcnl.wisc.edu/publications/archive/184.pdf); and for a working model: Harm, M. W., & Seidenberg, M. S. (2004). Computing the Meanings of Words in Reading: Cooperative Division of Labor Between Visual and Phonological Processes. Psychological Review, 111, 662-720 (http://lcnl.wisc.edu/publications/archive/43.pdf).

My view on what is a word and how it is perceived is shortly presented in my parallel topic "What is a unit of reading?".

In older models there was some mixture of notions, to my mind, leading to wrong architecture and wrong conclusions.

As I said before, asking these kinds of "either or" questions on such seemingly simple "things" as the way lexical items are stored in the human lexicon and the experiments on them, is relatively predictable: "Well, it depends upon which levels and which models and which experiments one is dwelling upon. Essentially, every response here to that question supports what I am saying. Think about it. Philosophically, I don't even think science is and "either or" thing at all. Here is a good either or question. Does the brain control the mind or does the mind control the brain? Either or!!!

Hugh Byuckinghamn

There is no mind without a brain. The brain starts working without a mind since the very beginning! And this is not even philosophy! The mind controls the Brain? How have you imagined this?

No, what I mean is that the software governs much of the functioning of the hardware.

Mental states are natural objects and it is most often the case that mental states are early agents of behavior. I just play around with "mind/brain" because I don't believe the hyppocampus remembers anything; the owner of the brain where the hyppocampus rests if surely the one who remembers. Only by a fanciful mapping of form to meaning and a "category mistake" do we claim that it is the brain and not he who has the brain does anything buy "dumb" cognition. I like the phrase "dumb cognition" because it is cognition out of awares.

You might now be calling ME dumb, but I stick to this until reductionism in science rears its ugly head. I have always liked the pineal gland and its splendid metaphors for reaching the ceiling of the soul.

Hugh Buckingham

I think, "either or" is an inherent human question, it is our permanent problem of choice, which we have to solve every step of our life way, in science as well. Your comment also seems like containing this "either or" question: is human cognition (and the whole world, maybe) materialistic or idealistic? I think, every mind/brain researcher starts as a materialist, but after a long science career becomes a bit (or more) idealistic. It is normal, because we're trying to learn mind by mind itself, always staying within the subject without being able to look at it from outside. I'm a bit agnostic here.

As for my title question, it means if one more human science notion is simple or sophisticated. I'm not adherent to simplicity.

As I do not profess any religion, I am a " atheleologist" , I do not discuss matters such as those or even about any kind of soul. This way I prefer to leave the discussion

No religion, no politics, just science. It's my scientific position that any object cannot be understood from within itself. In Chinese words "subjectively" and "objectively" in direct translation mean "master's view" (主观) and "guest's view" (客观) respectively. We need this "guest's view", a look from outside, to be objective. In case of our mind it is impossible, so we have limits in learning it. But I believe, we are still far from these limits, there's enough place for us to study.

Dmitry:

As long as Obama is RE-elected and gets four more years, there will BE no science - only creationism and other cokamaymee dreck spun out by the catholics and the mornorms, along with the blithering idiot tea baggers. Science? They have NONE of it. That is the stuff of Chuck Darwin and Ayn Rand. Oops! Paul Ryan(d) is an Ayn Rand wannabe, who is not even aware that Ayn Rand was born in Russia, under the name of Alissa Rosenbaum, 1905, St. Petersburg, Russia.

What a brilliant conversation between a couple of geniuses that I hated seeing derailed at the end. I hope that Dmitry and Osmar will reengage.