"... In PAGE 2: ... We presented several optimization models as examples of powerful tools for studying the design of human-computer interaction in information foraging. FRAME OF ANALYSIS Table1 presents a summary of an overall framework for studying human-computer interaction from an ecological and cognitive perspective. I should point out that all of the pieces may be found in existing literature in recent evolutionary theory, cognitive science, and measurement, so no claim for novelty is being made.... In PAGE 2: ... It is useful, however, to review the big picture. Table1 presents four levels of analysis: (1) an adaptive analysis level where it is assumed that the structure of behavior can be understood in terms of its adaptive fit to the structure and constraints of the environment, (2) a knowledge level where descriptions of observed behavior assume that it is the product of human purposes and knowledge (i.e.... ..."

"... In PAGE 1: ... a finite-state grammar or a dialog control table. In Table1 , the system prompts present a sequence of questions in an attempt to solicit more constraints from the user, with the goal of obtaining a small data subset from the database. The system responses are generated from a set of rules that only anticipate one of a handful of situations: (1) when the set of en- tries returned is too large, (2) the set of entries is adequately small to enumerate, and (3) no available entries have been returned.... ..."

"... In PAGE 3: ... These issues may be posed as five questions that a system user must be able to answer to accomplish some action. Table1 shows how each question has a familiar GUI answer. Further, each one poses some challenges that are easily solved by sticking to the existing GUI paradigm and its simpler hand-held counterparts.... ..."

"... In PAGE 6: ...will adapt that model, it is appropriate to provide some details. See Table3 , based on the Table of Contents of [ACMS92], which shows the various sections of the curricular document. It is noteworthy that there are a number of types of overlap between HCI and MHIA issues.... ..."

"... In PAGE 5: ...1), after regressing out pretest. 4. Human-Computer Results Table1 presents our best results on correlations between Dialogue Act Measures and learning in our human-computer corpus. The rst column lists the measure (total (#), percentage (%) or ratio (Rat:) of the Dialogue Act per student).... In PAGE 5: ... The quantity of tutor questions also relates to student learning, as evidenced by the strong positive correlation between the overall percentage of all tutor Question Acts and learning. Table1 also shows a slight trend for tutor Positive Feedback to positively correlate with learning. Other studies have shown positive relationships between encouragement during computer tutoring and student outcomes [14].... In PAGE 6: ... 5. Human-Human Results Table 2 presents our best results on correlations between Dialogue Act Measures and learning in our human-human corpus, using the same format as Table1 . As shown, the type of dialogue acts used by students relates to how much students learn in our human- human corpus too.... ..."

"... In PAGE 12: ... The view advanced here, which is intended to found a computational science on a scienti c basis, contrasts with the orthodox view of the computational sciences depicted previously in Table 1. Table4 exempli es the new view, which is an elaborated (and transposed) version of Table 2. The new view calls for developing a descriptive taxonomy of scienti c activity.... In PAGE 15: ... Then the boundaries between the enterprise of science as a whole (the acquisition and organization of the knowledge of the world) and AI (the understanding of how knowledge is acquired and organized) will become increasingly fuzzy. We regard the proposal of this paper as entirely consonant with Newell apos;s perception, although we have added the concept of generic tasks that cut across the sciences as in Table4 . Newell apos;s vision of building computational systems to do \object-level quot; science corresponds to our emphasis on science automation; our view of computational science as task-centered and discipline-generic calls at least for adjustments to the disciplinary structure of science, if not for Newell apos;s transformations.... ..."

"... In PAGE 5: ... Other potential change dimensions include: cost estimates, time estimates, and level of complexity. Table2 shows a COR matrix constructed with three levels of system automation and the entity sets derived above. The set of altered entities is further subdivided to reflect their appropriate categorization in the old and new systems.... ..."

"... In PAGE 6: ...wo design phase artifacts (i.e., data flow diagrams). Table3 is the detailed COR matrix that results from analyzing the certificate processing DFDs with respect to automation categories. Table 4 presents the change measures for the reengineered certificate handling process.... In PAGE 7: ...84 Previous Level of Automation 0.28 How can we interpret these measures? The COR matrix in Table3 shows us that a fairly large set of entities are in the core set, existing in both the old and new systems. However, the altered/unaltered dichotomy shows that many of these processes are modified with respect to their automation category.... ..."