DESIGN PRINCIPLES:

Pcues or PQs - The "P" stands for  'phonic', 'phonemic', 'pronunciation' and ‘parallel’. Phonic because they cue sounds, phonemic because they cue sub-sound boundaries, pronunciation because its not knowledge of relations that the cues prompt, its how to pronounce the sounds and parallel because they create a parallel process path for decoding to draw upon.

Ambiguity Reduction – The primary function of the Pcue system is to reduce the letter-sound correspondence ambiguity that is inherent in the ‘code’ and problematic to the process of learning to read. 

Alphabet and Spelling Independent – Pcues do not require any changes to the alphabet nor do they require any changes to spelling. Pcues are variations in the appearance of letters that cue readers to pronounce letters in ways that corresponds to the subsounds of words..

Readway Signs and Phonic Serifs  - Pcues are analogous to ‘highway signs’ that inform a driver’s movement in traffic. Pcues are ‘readway signs’ that inform a reader’s pronunciation while reading. Just as serifs are intended to guide the eye toward more fluid letter-appearance recognition, Pcues are intended to guide the mind toward more fluid letter-sound recognition.

Font General – Pronunciation Specific – Pcues are font-general not letter-specific. They, like ‘bold’, ‘Italics’ and ‘underline’, are general attributes of a font that can in principle be applied to any letter.  However, the way the cues are used to guide pronunciation is specific to a letter’s sound in the specific word it is encountered in.  Just as ‘bold’, ‘italics’ and ‘underline’ are used to indicate or emphasize a specific meaning for the word or words being read, Pcues are used to indicate or emphasize a specific pronunciation for the letter or letters being read.

Visual Distinction - Pues are visually distinct and easy to recognize. They do not obscure underlying letter recognition features.

Mental Resource Efficiency - Pcues are as intuitive as possible. To facilitate this the appearance of a cue is, everywhere possible, a morphic analogy of the letter sound quality it represents (i.e. larger for louder  - faintly visible for silent). By making the discernable visual distinctiveness analogous to a corresponding discernable distinctiveness in pronunciation, we minimize the memory and attentional processing resources required to both recognize and apply a cue.

Learning Ease - the pedagogy for learning to recognize and apply each cue is very simple and direct.

Minimal Number – The intent of Pcues is to reduce unnecessary mental processing operations. The trial-optimized system will achieve a balance between the numbers of cues, their ease of learning and the ambiguity the cues reduce. 

Minimal Distraction - the cues should pose little distraction to an experienced reader.

Paper Based – Throughout the foreseeable future, the overwhelming majority of people will learn to read on paper. Though Pcues can be used to enhance the use of electronic displays, their design has avoided even the use of color.  It is our intention that the ‘standard’ system be available to every beginning reader. Building on the standard system, a variety of additional features, from color to voice recognition based interactive coaching, can be developed for those with access to a learning appliance or computer.

Training Wheels – Pcues act like training wheels by helping to keep readers from falling out of flow. Like training wheels, they are designed to ‘come off’ when the reader is ready. Our assumption is that because the children will be learning to read with the normal alphabet and spelling (just getting help from the cues in decoding how they 'sound' together) their repeated successes with ever more familiar sub-word decodings will give them a  'training wheels' effect. When we later phase out or drop the cues, the majority of their reading experience will still be applicable. Recognizing letter combinations/sub-word sounds and how they combine should be much easier (very closely paralleling the bicycle training wheels metaphor).

More On via Questions and Answers …

Adams (1990) reported that when she asked her University students to read words made up of letters of different fonts and sizes, the students were still able to read them fairly easily. Students in Adams’ study also reported that when the words were presented very quickly, the letters all appeared, in their minds, to be of similar size and font. This suggested to her that readers very quickly transform what they see into more abstract letter forms.
             From Reading Difficulties - University of Auckland

KINDS OF CUES:

(LN) Letter Name Sounds: One of the most difficult confusions for developing readers to resolve is an unintended consequence of learning the ABCs (letter name sounds). Because we teach children the ‘ABCs’ before we teach them to read, their minds become conditioned to associate a letter with its letter name sound.  When later learning to read, the child’s response to seeing a letter is to ‘hear’ its letter name. As it is seldom the case that a letter sounds like its letter name, the ambiguity created by this association works against the process of learning to read. We must provide beginning readers a way to determine when a letter’s sound is to be read as its letter name and when it is not. 

Letter name sounds sound more sharply distinct and 'louder' than non-letter name sounds. There is a direct analogy of form between their discernable sound difference and rendering their letters both larger and more boldly distinct.  By using Larger Bold to indicate letter name sounds the meaning of the visual cue is minimally abstract - more intuitively obvious - and therefore easier to remember.

(CL) Combined Letter Sounds (th, ph, ch, sh, etc..):  Combined letter sounds are in a class by themselves. The problem with combined letter sounds is recognizing that their individual letters are not to be decoded separately but combined to represent their distinct assigned sounds. Condensed spacing can be used to cue readers to recognize such groups and to indicate that they are to make one sound. This immediately removes the blended letters from consideration for isolated decoding.

The spacing in the ‘ch’ in ‘change’, the ‘th’ in ‘the’ and the ‘ph’ in ‘phoneme’ are in contact with one another to cue that they are letter combinations to be read as one unit. Placing letters in contact with one another to indicate that they are to be read as a group is a perfect morphic analogy.

Alternate Letter Sounds:  The most frequently occurring confusion is which of a letter's alternative sounds (non LN sounds) is a particular letter being read supposed to sound like? 

There are a number of alternate letter sound distinctions that we can leverage in cue design:

Silence – some letters are not pronounced as in the case of the ‘a’ in ‘sea’ and the ‘k’ and ‘w’ in know.

1^st Alternate – some letters only have two sounds. Of the letters with multiple sounds there is often a most common, non-LN, sound, such as the ‘ae’ sound of ‘a’.

Discreteness – some letters have letter sounds that do not sound at all similar as in the case with the letter ‘c’, which can also sound like the totally different sound ‘k’.

Spectrum – some letters have alternative letter sounds that all sound similar in some way as in the case with the letter ‘a’, which in addition to the LN ‘a’ sound can also sound like aw (talk), or ae (dad). In this sense, ‘a’, ‘ae’ and ‘aw’ are variations along a ‘spectrum’. 

(SL) Silent / Unpronounced (possibly minimally pronounced) Letter Sounds: GRAYED OUT represents a minimally abstract and maximally analogous, intuitive and easy to remember way of cueing:

(LNs for whole words are problematic and perhaps only usable at the beginning stage of TWFL.)

With the Letter Name (LN), Combined (CL) and Silent (SL) cues clearing the reading stream of the simpler ambiguities we can address the more complex problems associated with the remaining alternate letter sounds.

(AL1) 1st Alternate Letter Sound  - With LN cues enabling readers to immediately determine whether a questionable letter's sound is or is not its letter name (LN) it becomes possible to allow the normal presentation of the letter to represent the most frequently used non-letter name sound (AL1). This approach is not an added cue, it is a new definition of the default sound of a non-cued letter.

In 'mightily' the first 'i' is the LN 'i' and is therefore LN cued. The second 'i' is 'eh', the 'i's most common non-LN sound (AL1) and is therefore plainly rendered.  The first 'e' in 'remake; is the LN 'e' sound and is therefore LN cued. The second 'e' is the nearly silent 'eh', the 'e's most common non-LN sound (AL1) and therefore plainly rendered.  In 'imagination' the first 'a' is 'ah', the 'a's most common non-LN sound (AL1) and is therefore plainly rendered, the second 'a' is the LN 'a' and is therefore LN cued. (This approach would work well for developing readers who started out learning to read this way. In the case of remediation, where learners have developed confused associations, one of the other modes of drawing attention to the alternate sounds will probably work better).

It could be argued that we should stop right here. By using only the simple cues discussed thus far, and teaching developing readers to rely upon them, the ambiguity involved in decoding can be dramatically reduced. This may be the right balance of cue complexity to ambiguity reduction. Only trial-based research will tell.  

(DL) Alternate Letter Sounds – Discrete:  Many letters are used to represent sounds that have no resemblance to their letter name sound (‘c’ as ‘k’, ‘s’ as ‘c’, ‘x’ as ‘z’, etc.). One approach to cueing is to draw upon their difference in tone.  Each different letter sound can be distinguished as being either lower or higher in tone or pitch than the letter’s LN sound. Using this basis for discernment we could use rotation or elevation as a cue for prompting the reader to know that this letter has discrete letter sounds as opposed to a spectrum of letter sounds and, subsequently, which of the letters alternate sounds it is to make.

The ‘c’ in ‘can’ is a ‘k’ sound that is lower in tone than it’s LN and can be represented by rotating it backwards or lowering it. The ‘s’ in ‘sea’ is a ‘c’ sound and can be represented by rotating it forward or raising it. Vertical centering or elevation represents the more visually-conceptually analogous rendering but presents line spacing issues. Rotation is less morphically analogous but seems easier on the eye and creates less formatting problems.

(SP) Alternate Letter Sounds – Spectrum: As in the above example, each spectrum letter’s (‘a’, ‘ae’, ‘aw’, etc.) alternate sounds can be represented on a scale within which each alternate sound is either ‘lower’ or ‘higher’ in tone/pitch than its letter name sound. For example, the 'i' in 'animal' sounds like 'eh' which is a 'lower' tone then the 'i' in 'his' which sounds like 'ih'.   

The ‘a’ in ‘had’ is lower in tone than it’s LN and can be represented by rotating it backwards or lowering it. The ‘a’ in ‘walk’ is even lower in tone than the ‘a’ in ‘had’ and can is represented by a greater exaggeration of rotation or lowering.

(DUR) Duration: Another distinction available to letters that have a spectrum of alternate sounds is the difference in length of time, or duration, that each alternate sound lasts. In this case we could 'stretch' the width of characters to emphasize their longer or shorter durations.

In cake the ‘a’ sound is it LN and therefore LN cued. In ‘had’ the ‘a’ sound is the shorter ‘ae’ sound and is therefore cued by its slim width. In ‘walk’ the ‘a’ sound’s duration is longer and is therefore cued by its stretched width.

Integrated Alternate Letter Sound Systems:

One method would be to use rotation for spectrum cues and elevation for discreet cues:

Another alternative is to use duration for spectrum cues and elevation for discreet cues:

(PM) PhoniMorphs: Another method would be to combine stretch or shrinkage for duration with a directional skew representative of higher or lower pitch:

It is also possible to integrate AL1 and sound pitch based cues into a common system. In this case the rotation (or duration or Phonimorphed) cues would only be used for the not most common non-letter name sounds and elevation would be used for the not most common discreet sounds. This actually uses less cues to provide more distinction.

In the above example, the 'y' in 'Herby' is making an 'e' sound, the most common non LN sound (AL1) for a 'y' therefore it is left plain (it could also be rendered a silent sound with the b rendered as the LN 'b'). In 'gym' the 'y' is making and 'eh' sound which is lower in tone than the LN 'y' and is therefore lowered. The 'y' in 'nearby' is making an 'i' sound which is higher in tone than the LN 'y' and is therefore elevated. In the case of the 'o' in 'on' it is making the most common non LN 'o' sound (AL1) and is therefore left plain. The 'o' in 'to' is making the 'ooh' which is higher in tone than the LN 'O' and is therefore rotated right.

(SG) Segmentation: to support the development of Phonemic Awareness and to avoid the decoding assembly problems posed by 'longer' words we can use extended spacing to cue phonemic / syllabic distinctions:

OTHER POSSIBLE CUES:

(AMP) Amplitude: Another dimension of distinction in heard or virtually heard language is the spectrum of amplitude (loudness to relative softness). This is the varying 'loudness' that is 'heard' in the unfolding stream of sound we hear as words.  The ideal form-analogous cue for this discernable variation in amplitude is SIZE. (LN)s are almost always 'loud' (why we favor the (LN) cue distinction be larger and bolder).

(OUT) Outlined: As an alternative to rotation, elevation or widened or phonimorphic cues, we could use outline letters and partially fill the outline to suggest the tone or duration difference. This can be done using left to right or up and down for tone scale or duration. (shown extra large for viewing on a low resolution display):

OUR STARTING SYSTEM:

The final visual variation styles for the cues should be the result of a collaborative effort which includes reading specialists, graphic artists, font designers and, of course, extensive learning and testing with developing readers. However, of the cues described here, these are our starting set:

Letter Name (LN), Combined Letter (CL), Silent (SL) and Segmentation (SG) cues are the core set.  They all offer significant ambiguity reduction, are easy to recognize and appear as visual analogs of the pronunciation directions they cue.

Of the more complicated remaining cues, we favor elevation for discrete letter cues (DL) and an evolved version of PhoniMorphs for spectrum letter cues (SP). First alternate (AL1) cues should be used only when the system is introduced from the beginning of the learning to read process.

Our implementation examples are incomplete and not fully consistent. We are not suggesting that these embodiments of the concept are the ‘right’ ones. To determine which variations of character presentation are ‘right’ for each class and kind of ambiguity will require research and testing.  The thresholds of visual recognition for each kind of cue must be researched. In particular, AL cues, which require more post-recognition processing, need to evolve via trials with beginning readers.

What is right is the overall concept. With it we can iterate our way to a system of cues that will dramatically reduce the ambiguity that impedes learning to read. We can do this without change to the alphabet or the English system of spelling and without having to train the minds of our students in any way approaching the complexity of phonics. We can create a tool that can scan through the text of any document and automatically reformat the appearance of the document (by employing the cue system) and consequently render it radically easier to read for people learning to read.  

©Copyright 2001 - 2003: Training Wheels for Literacy & Implicity