The data for the marginally literate men (MLM) again illuminate differences between languaging and conceptualizing processes. At the slowest rate of presentation, 128 wpm, these men were able to aud and conceptualize almost as well as the college students, as indicated by their immediate retention scores. Their tracking scores, on the other hand, are almost 40% below the college students, indicating that the marginally literate men (MLM) could not perform the Sequence B activities very well-which is reasonable since these activities are reading activities, and these men are not skilled readers (ie., decoders of print to internal spoken form). However, we note also that the MLM have difficulty with the Sequence A processes at the faster speech rates. They are unable to conceptualize the information even by auding as well as the college students do when faster speech rates are used (again, we assume from the very low tracking scores, which are almost always errors of omission rather than commission, that the MLM are dropping reading processes in favor of auding processes; this seems reasonable because of the low reading skills-fifth-grade level-of these men, and because the auding message sets the pace).

The foregoing analysis, although it is obviously incomplete and fragmented (What are the subprocesses in Sequence A? How is the interface between languaging and conceptualizing processes to be construed?), serves to illustrate the distinction we are making between languaging and conceptualizing. More to the point of Hypothesis 3, however, the arguments and studies cited above tend to support the notion that reading utilizes the same languaging and conceptualizing skills used in auding. Limitations in the rates at which languaging and conceptualizing can be performed place upper limits on both auding and reading.

Perhaps conceptualizing ability can be improved through education and training in reasoning. Languaging ability seems limited by physiological factors involved in articulating. Hence training in rapid auding or reading will most likely have to focus on training the auder or reader to ignore much of what he hears or sees, and to sample messages through scanning and skimming techniques using cues such as pauses and inflection for rapid scanning of speeded speech, and italics, indentations, and underlinings for scanning print displays.

While languaging is obviously involved in such processes whenever the ear or eye fix on a segment of a message, the conceptualization processes must play the major role in synthesizing a meaningful story out of fragments of the message. If such rapid conceptualizing is to be accomplished, a person must have a plan for attacking the materials, and this plan will be more faithfully executed when the display is under the control of the scanner. Thus, the reading display lends itself to such rapid scanning, and this may be what happens when we hear of "reading" at 3,000, 4,000, or even 1 million (!) words per minute. But it is clearly not what the "typical" reader does in "typical" situations, and hence the study of such processes is beyond the scope of the present review.