Báo cáo khoa học: "A Linear-time Model of Language Production: some psychological implications" doc

A Linear-time Model of Language Production: some psychological implications extended abstract David D.. This is a serious problem for ATN-b~sed theories of production since they have n

A Linear-time Model of Language Production: some psychological implications

(extended abstract) David D McDonald MIT A r t i f i c i a l I n t e l l i g e n c e L a b o r a t o r y Cambridge, M a s s a c h u s e t t s

Traditional psycholinguistic studies of l a n g u a g e

p r o d u c t i o n , u s i n g e v i d e n c e f r o m n a t u r a l l y o c c u r r i n g

e r r o r s i n s p e e c h [ 1 ] [ 2 ] a n d f r o m r e a l - t i m e s t u d i e s o f

h e s i t a t i o n s a n d r e a c t i o n t i m e [ 3 ] [ 4 ] h a v e r e s u l t e d in

m o d e l s o f t h e l e v e l s a t w h i c h d i f f e r e n t l i n g u i s t i c u n i t s

a r e r e p r e s e n t e d a n d t h e c o n s t r a i n t s o n t h e i r scope

T h i s k i n d o f e v i d e n c e b y i t s e l f , h o w e v e r , can t e l l us

n o t h i n g a b o u t t h e c h a r a c t e r o f t h e p r o c e s s t h a t

m a n i p u l a t e s t h e s e u n i t s , as t h e r e a r e m a n y a p r i o r i

a l t e r n a t i v e c o m p u t a t i o n a l d e v i c e s t h a t a r e e q u a l l y

c a p a b l e o f i m p l e m e n t i n g t h e o b s e r v e d b e h a v i o r It w i l l

b e t h e t h e s i s o f t h i s p a p e r t h a t i f p r i n c i p l e d , n o n -

t r i v i a l m o d e l s o f t h e l a n g u a g e p r o d u c t i o n p r o c e s s a r e

t o b e c o n s t r u c t e d , t h e y m u s t be i n f o r m e d b y

c o m p u t a t i o n a l l y m o t i v a t e d c o n s t r a i n t s In p a r t i c u l a r

t h e d e s i g n u n d e r l y i n g t h e l i n g u i s t i c c o m p o n e n t I h a v e

d e v e l o p e d ("MUMBLE p r e v i o u s l y r e p o r t e d in [ 5 ] [ 6 ] )

is being investigated as a candidate set of such

c o n s t r a i n t s

A n y c o m p u t a t i o n a l t h e o r y o f p r o d u c t i o n t h a t is t o

be interesting as a psycholinguistic model m u s t meet

certain m i n i m a l criteria:

( 1 ) P r o d u c i n g u t t e r a n c e s i n c r e m e n t a l l y , in t h e i r

n o r m a l l e f t - t o - r i g h t o r d e r , a n d w i t h a w e l l -

d e f i n e d " p o i n t - o f - n o - r e t u r n " s i n c e w o r d s

o n c e said can not be invisibly taken back~

(2) M a k i n g the transition from the n o n -

linguistic "message"-level representation to

the utterance via a linguistically structured

buffer of only" limited size: people are not

capable of linguistic precognition and can

I This report describes research done at the Artificial

Intelligence Laboratory of the Massachusetts Institute of

Technology Support for the laboratory's artificial

intelligence research is provided in part by the Advanced

Research Projects Agency of the Department of Defence

u n d e r Office of Naval Research c o n t r a c t

N 0 0 0 1 4 - 7 5 - C - 0 6 4 3

readily "talk themselves into a corner ''z (3) G r a m m a t i c a l robustness: people m a k e v e r y

f e w grammatical errors as compared w i t h lexical selection or planning errors ("false

s t a r t s " ) [ 7 ]

T h e o r i e s w h i c h i n c o r p o r a t e t h e s e p r o p e r t i e s as a n

i n e v i t a b l e c o n s e q u e n c e o f i n d e p e n d e n t l y m o t i v a t e d

s t r u c t u r a l p r o p e r t i e s w i l l be m o r e h i g h l y v a l u e d t h a n

t h o s e w h i c h o n l y s t i p u l a t e t h e m

T h e design incorporated in M U M B L E has all of these properties~ they follow from t w o k e y

i n t e r t w i n e d stipulations hypotheses motivated b y intrinsic differences in the kinds of decisions m a d e

d u r i n g language production and b y the need for an efficient representation of the information on w h i c h the decisions d e p e n d (see [8] for elaboration)

(i)

(~)

T h e e x e c u t i o n t i m e o f t h e p r o c e s s is l i n e a r in

t h e n u m b e r o f e l e m e n z s in ~he i n p u t

m e s s a g e , i.e the realization decision for each

e l e m e n t is m a d e only once and m a y not be revised

T h e representation for pending realization decisions a n d planned linguistic actions (the results of earlier decisions) is a surface-level syntactic phrase structure a u g m e n t e d b y explicit labelings for its constituent positions (hereafter referred to as the tree) 3

This working-structure is used simultaniously for control (determining

w h a t action to take next), for specifying constraints ( w h a t choices of actions are

Z In addition, one inescapable conclusion of the research

on speech-errors is that the linguistic representation(s) used d u r i n g the production process must be capable of

r e p r e s e n t i n g positions independently of the units (lexical or phonetic) that occupy them This is a serious problem for ATN-b~sed theories of production since they have no representation for linguistic structures that is independent front t h e i r representation of the state of the process

message elements These are replaced by syntactic/lexical structures as the tree is refined in a top-down, left-to-right traversaL Words are produced as they are reached at (new) leaves, and grammatical actions are taken

as directed by the annotation on the traversed regions

ruled out because of earlier decisions), for

the representation of linguistic context, and

for the implementation of actions motivated

only by grammatical convention (e.g

a g r e e m e n t , w o r d - o r d a r w i t h i n t h e clause,

m o r p h o l o g i c a l specializations; see [6])

T h e r e q u i r e m e n t of l i n e a r t i m e r u l e s o u t a n y

d e c i s i o n - m a k i n g t e c h n i q u e s t h a t w o u l d r e q u i r e

a r b i t r a r y s c a n n i n g o f e i t h e r message o r tree Its

c o r o l l a r y , " I n d e l i b i l i t y " , 4 r e q u i r e s t h a t message be

r e a l i z e d i n c r e m e n t a l l y a c c o r d i n g to t h e r e l a t i v e

i m p o r t a n c e o f t h e s p e a k e r ' s i n t e n t i o n s The p a p e r w i l l

d i s c u s s h o w as a c o n s e q u e n c e o f t h e s e p r o p e r t i e s

d e c i s i o n - m a k i n g is f o r c e d t o t a k e place w i t h i n a k i n d

o f blinders: r e s t r i c t i o n s o n t h e i n f o r m a t i o n a v a i l a b l e

f o r d e c i a l o n - m a k i n g and on t h e possibtUtias f o r

m o n i t o r i n g a n d f o r i n v i s i b l e s e l f - r e p a i r , all d e s c r i b a b l e

i n t e r m s o f t h e u s u a l l i n g u i s t i c v o c a b u l a r y A f u r t h e r

c o n s e q u e n c e is t h e a d o p t i o n of a "lexicalist" p o s i t i o n o n

t r a n s f o r m a t i o n s (see [ 9 ] ) , i.e once a s y n t a c t i c

c o n s t r u c t i o n h a s b e e n i n s t a n t i a t e d in t h e tree, t h e

r e l a t i v e p o s i t i o n o f i t s c o n s t i t u e n t s c a n n o t be modified;

t h e r e f o r e a n y "transformations" that apply must do so

a t t h e m o m e n t the c o n s t r u c t i o n is instantiatad and on

t h e b a s i s o f o n l y t h e i n f o r m a t i o n available at t h a t time

T h i s is b e c a u s e t h e t r e e is n o t b u f f e r of objects, b u t a

p r o g r a m o f s c h e d u l e d e v e n t s

N o t i c e d r e g u l a r i t i e s in s p e e c h - e r r o r s h a v e

c o u n t e r - p a r t s i n MUMBLE's d e s i g n 5 w h i c h , to t h e

e x t e n t t h a t i t is I n d e p e n d e n t l y m o t i v a t e d , m a y p r o v i d e

a n e x p l a n a t i o n f o r t h e m One e x a m p l e is t h e

4 I.e decisions are not subJeCt to backup-="they are

~rritten in indelible ink" This is also a property of

Marcus's "deterministic" parser It is intriguing to

speculate that indelibility may be a key characteristic of

psychologically plausible performance theories of natural

language

5 MUMBLE produces t e x t not speech Consequently it

has no Knowledge of syllable structure or intonation and

can make no specific contribution= to the explanation of

errors at that level

errors w h e r e functional morphemes such as plural or tense are "stranded" at their ori~inal positions, e.g

" M y l o c a l s a r e m o r e v a r i a b l e t h a n t h a t "

I n t e n d e d - " v a r i a b l e s a r e m o r e l o c a l "

" W h y don't w e E o to the 24hr Star M a r k e d and

y o u can see m y friend check in E cashes."

I n t e n d e d : " cashing checks."

O n e o f t h e t h i n g s t o be e x p l a i n e d a b o u t t h e s e e r r o r s is

w h y t h e t w o c l a s s e s o f m o r p h e m e s are d i s t i n g u i s h e d - -

w h y d o e s t h e " e x c h a n g i n g m e c h a n i s m " e f f e c t t h e o n e

a n d n o t t h e o t h e r ? The f o r m of t h e a n s w e r t o t h i s

q u e s t i o n is g e n e r a l l y a g r e e d upon: t w o i n d e p e n d e n t

r e p r e s e n t a t i o n s a r e b e i n g m a n i p u l a t e d a n d t h e

m e c h a n i s m a p p l i e s t o o n l y o n e of t h e m MUMBLE

a l r e a d y e m p l o y s t w o r e p r e s e n t a t i o n s of r o u g h l y t h e

c o r r e c t d i s t r i b u t i o n , n a m e l y t h e p h r a s e s t r u c t u r e t r e e ( d e f i n i n g p o s i t i o n s a n d g r a m m a t i c a l p r o p e r t i e s ) a n d

t h e m e s s a g e ( w h o s e e l e m e n t s o c c u p y t h e p o s i t i o n s a n d

p r o m p t t h e s e l e c t i o n o f w o r d s ) By i n c o r p o r a t i n g

s p e c i f i c e v i d e n c e f r o m s p e e c h - e r r o r s i n t o MUMBLE's

f r a m e w o r k ( s u c h as w h e t h e r t h e q u a n t i f i e r all

p a r t i c i p a t e s i n e x c h a n g e s ) , it is possible to p e r f o r m

s y n t h e t i c e x p e r i m e n t s t o e x p l o r e t h e impact of s u c h a

h y p o t h e s i s o n o t h e r a s p e c t s of t h e design The

i n t e r a c t i o n w i t h p s y c h o l i n g u i s t i o s t h u s becomes a

t w o - w a y street

The f u l l paper 6 w i l l develop the n o t i o n of a

l i n e a r - t i m e p r o d u c t i o n process: h o w i t is accomplished

a n d t h e s p e c i f i c l i m i t a t i o n s t h a t it imposes, and w i l l

e x p l o r e i t s i m p l i c a t i o n s as a p o t e n t i a l e x p l a n a t i o n f o r

c e r t a i n c l a s s e s o f s p e e c h - e r r o r s , c e r t a i n h e s i t a t i o n a n d

s e l f - c o r r e c t i o n data a n d c e r t a i n l i n g u i s t i c constra_nts

6 Regretably, the completion of this paper has been delayed in order for the author to give priority to his dissertatlon

References

[I] Garrett M.F (1979) "Levels of Processing in Sentence Production", in Butterworth ed

L a n g u a g e Production V o l u m e I, Academic Press [2] Shattuck Hufnagel, S (1975) Speech Errors a n d

S e n t e n c e Production P h D Dissertation, Department of Psycholog~v, MIT

['3] Ford M & Holmes V.M (1978) "Planning units and syntax in sentence production", Cognition 6, 35-

63

['4] Ford M (1979) "Sentence Planning Units: Implications for the speaker's representation of meaningful relations underlying sentences", Occasional Paper 2, Center for Cognitive Science, MIT

['5] McDonald, D , D (1978) "Making subsequent references., syntactic and rhetorical constraints", TINLAP-g University of Illinois

[ 6 ] (1978) "Language generation: Automatic Control of Grammatical Detail", COLING-

78 Bergen Norway

['7] Fay, D (1977) "Transformational Errors" International Congress of Linguistics Vienna, Austria

[8] M c D o n a l d D.D (in preparation) Natural Language Production as a Process of Decision-making

U n d e r ConsU'alnt Ph.D Dissertation, Department

of Electrical Engineering and Computer Science, MIT

[9] Bresnan, J (1978) "Toward a realistic theory of grammar", in Bresnan Miller, & Halle ads

L i n g u i s t i c T h e o r y and Psychological Reality Mrr Press