PSY2011 - WUG Introduction Theoretical Background

These pages are demonstration pages of mixed mode deli very for an on-campus course. The curriculum and course structure has changed substantially for 1998, but these pages have been retained here as an example of a particular use of the Web in teaching.

Introduction to the Language Acquisition Experiment
The WUG Lab

The introduction to this lab contained much material as you all know, hence the duration of the introductory session. The Language Acquisition web page contains details about the elicitation task, stimulus items, notes on testing children and data recording information. This summary is an overview of the psycholinguistic content discussed in the first part of the introductory session. In actual fact, it is a copy of the overheads presented last week with a few additional notes.

Morphemes, allomorphs and morphophonemic rules

Morphemes

Morpheme refers to a distinct unit of meaning with a specified sound pattern. It is the smallest unit of sound that has distinct meaning. A single word may contain one or more morphemes, for example;

  cat			1 morpheme
  catastrophe 	 	1 morpheme
   	(this word cannot be broken down into smaller
    	meaningful parts with respect to the original 
    	word. That is the /cat/ in catastrophe does not 
    	have the same meaning as the (pussy)/cat/ above)
  cats  			2 morphemes 
	(cat is the root and -s is the plural morpheme)

Morphemes and syllables have quite different meanings although they are both sub-units of words. Morphemes and phonemes are also different, morphemes referring to meaning and phonemes referring to sound.

Types of Morphemes

Bound morphemes need to be attached to another morpheme in order to be meaningful. For example the morpheme -s in cats denoting the plural form does not have independent meaning by itself.

Examples of free morphemes include prepositions such as the, a, in, and that. These morphemes are known as function words. They don't actually have a meaning by themselves, despite being stand-alone, bona fide words. Whether a given morpheme is free or bound depends on the specific language.

There is also a distinction between root morphemes and affixes.

	prefixes 		eg. unhappy, distaste
	infixes		(not used in English aside from the "emphatic register"
				eg. fan -........-tastic).
	suffixes		eg. boyish, friendly.

The root morphemes are single entity words such as happy, boy, friend, that are distinct and meaningful on their own.

Derivational morphemes are morphemes which change the properties of the word to which they are bound. In such cases, the grammatical class of the root word is changed.

	For example, when -ish is added to boy to form 
	boyish, a noun has changed to an adjective.
	
	When -ly is added to adjective boyish, the adverb, 
	boyishly is formed. The meaning of boyishly 
	can be derived from the root morpheme boy.

Inflectional morphemes do not change the grammatical class of the root word.

	Examples of this type of morpheme include
		cat/-s		still a noun, but now
				in the plural form.
		bat/-ted	still a verb, only the 
				time frame changed,
				from present to past tense.

Allomorphs

In this study we are interested in testing children's use of regular plural allomorphs, where an allomorph is a variant form of the same basic morpheme, which is dependent on the phonological environment.

	Examples of the regular plural morpheme in English are:
		  [ez] 	horses
		  [z] 	dogs
		  [s] 	cats

The allomorphs have three different phonological forms, that is they sound different depending on the sound environment at the end of the word to which they are being added.

Complex Plural

The neutral vowel schwa ("e" upside-down) is inserted when we pluralise "horse" (with a silent "e") since it is almost impossible to say hors-s phonologically. The epenthetic (inserted) vowel is required in order to say (plural horse) hors[ez].

The phonological environment of the root word, determines when the epenthesised vowel is required. The form of the regular plural is determined by phonological conditioning. Morphophonemic Rules describe the variant pronunciations of a morpheme in their phonological environments.

	eg. For words that end in sibilants, that is the class of sounds 
	that are hissing or hushing sounds (eg. bush, garage, quiz), 
	you add [ez] for the plural allomorph.

Orthography vs Phonology:
In the case of the plural of horse, we add an "s" to the orthographical representation of that word. However phonologically we add [ez]. In the orthographical representation of witch[-ez] we add an "e", together with the "s", so the phonological representation of witches approximates the orthographical form. The basic form of language is spoken language, and it is the phonology, not the orthography that is of interest for this experiment.

Simple Plural

When a word's final sound is not sibilant, the plural allomorph has no inserted vowel. The plural [-s/-z] sounds are determined by whether the preceding phonological environment is voiced or voiceless. A voiced sound is where you can feel your vocal chords vibrating as you say it, for example - cab, cad, love, call, bee. All vowels are voiced. For these types of voiced sounds, the correct plural allomorph is the -z sound. We could add [-ez] and say bag-ez, but it does not sound appropriate.

Voiceless sounds at the end of words, include the following situations, cap, cat, back. The distinction between for example b/p d/t and v/f is that the first of the pair is voiced while the second is voiceless. As whispering is essentially devoicing, a "b" becomes "p" in a whisper. For voiceless word endings, the plural allomorph is [-s]. So we have cat-s, cap-s etc.

For the purposes of our experiment you are not required to distinguish between the two non-sibilant endings, [-s] and [-z]. The are very difficult to distinguish for the phonetically-untrained ear of the psychologist, especially if your subject whispers his/her responses.

Past Tense allomorphs

The regular past tense formation obeys similar morphophonemic rules in determining the type of allomorph required to form the past tense. If a word ends with an alveolar stop, that is a [-t] or [-d] sound, as in melt or plod, then we insert the schwa [e] with the d, to say melt-[ed], or plod-[ed].

Again whether the phonological environment is voiced or voiceless determines the correct allomorph for words not ending in an alveolar stop. Voiced examples, include words like climb and play, where the past tense forms end with a [-d] morpheme eg. played and climbed. Note we don't pronounce the [e], although we write it, highlighting again the distinction between the orthography and phonology of words.

Words that are voiceless, kick, cap, cough, take the past tense allomorph [-t].

You will know the "correct response" for the nonsense words, because you (tacitly) know the rules, and you are competent speakers of the English language. Remember, the correct response in terms of our experiment is the response an adult speaker of the language would use.

Good luck and enjoy the experience of data collection. Don't forget that it should be worthy of a lab diary entry.

Chris Hughes

Language Acquisition | Labs | PSY2011 Home Page | Psychology | Monash University