Editing Fis phenomenon (section)

=== Perception ===
The phonological performance of children is predominantly consistent and predictable, leading to the generally accepted notion that their performance is governed by a set of rules, and it is not a result of random deviations. These rules are used to navigate from the surface form (adult pronunciation) to child pronunciation. This idea might help explain the occurrence of things such as the Fis Phenomenon.

There is evidence to support the idea that a child manipulates isomorphic adult representations of language. This evidence stems from three areas: 1) that a child has the ability to recognize disparities in the adult form that the child is unable to produce 2) that the child understands their own speech and 3) their grammatical and morphological tendencies.

The role of perception in the phonological performance of children is that their lexical representation of the adult form is first passed through the child's perceptual filter. Meaning that the adult pronunciation, or surface form, is not necessarily the form that is being affected by the child's phonological rules. There is a clear difference between the adult form and the child's mental representation. Barton (1976)<ref>BARTON, DAVID, 1976. The Role of Perception in the Acquisition of Phonology. University College, London dissertation.</ref> tested this hypothesis and the results largely supported it, though there were later requests for a “perceptual explanation”.

The most notable example, shown below, illustrates the perception of consonant clusters compared to the child's output. Clusters consisting of [+nasal] followed by a [+voice] or [-voice] consonant are perceived differently by children.

<blockquote>mend → mɛn meant → mɛt<ref name=":0">{{Cite book|last=Smith, N. V. (Neilson Voyne)|title=Acquiring phonology: a cross-generational case-study|date=17 April 2014 |isbn=978-1-107-66295-7|oclc=949367454}}</ref></blockquote>

The nasal before a voiced consonant is long and notable (me'''n'''d).

The nasal before an unvoiced consonant is indistinct, leaving the following consonant as the most notable of the cluster.

The child's mental representation is then converted by a small set of rules called Realization Rules, which are used to reach the final form, the child's pronunciation. An example of the implementation of Realization Rules is informally illustrated in the sample derivation below, where a child consistently produced ''squat'' as ''[gɔp]'':

<blockquote>/skwɒt/ → [skwɔp]	(harmonizing a coronal to a preceding labialized sequence /kw/)

[skwɔp] → [kwɔp]	(deleting pre-consonantal /s/)

[kwɔp] → [kɔp]		(deleting post-consonantal sonorants)

[kɔp] → [gɔp]		(neutralizing the voicing distinction)<ref name=":0" /></blockquote>