22 eltects on teaming. i made the explicit and clear claim that there were no learning benefits possible and urged that we not continue to waste eiiort on the question until a "new the cry" was developed, l intended to stimulate discussion and l was not disappointed. betore l describe the reactions however, the discus sion turns to a briet review ot the argument, the importont Aspects oi the Learning From Medici Argument My early articles (Clark, 1933, 1935a) claimed, in part, that media are "mere vehicles that deliver instruction but do not influence student achievement any more than the truck mat delivers our groceries causes changes in our nutrition" (1933, p. 445). The articles presented evidence in support oi the hypothesis that instructional methods had been contounded with media and that it is methods which influs ence learning. Further, l claimed, that any nee essary teaclung method could be designed into a variety oi media presentations, 1 also ques- tioned the unique contributions at media attri- butes. Gavnel Salomon and others (Salomon, 1979) had argued that it was not the medium which influenced learning but instead certain attributes ot media that can be modeled by learners and can shape ttie development oi unique “cognitive processest" Examples ot media attributes are the mpacity ot television and movies to "zoom” into detail or to ~unwrap~ Khmerdimensional objects into two dimensions. The problem with the media attri- bute argument is that there is strong evidence that many very ditterent media attnbutce accomplish the same learning goal (tor exam- ple, there are a variety oi equally ettective ways to highlight details other than zooming). in every attempt to replicate the published media attribute studies (see studies cited by Clark, 1985c; Clark at Sugrue, 1968), a number oi very diilerent media attributes served the same or similar cognitive iuncuons. This point is critical to my argument. it there is no single media attribute that serves a unique cogn 've etiect ior some learning taslc, then the attri- butes must be proxies lor some other variables that are instrumental in learning gains. Emw Voumto 2 A Replaceabtmv Challenge it may be usetul to apply the following ”amt- chair experimental criteria” to any situation where it appears that media or attributes oi media have been instrumental in fastering learning gains: We need to ask whether there are olhzr media or analhn set a; media attributes that would yreiit stvitlrrtiiarriirig gains The question is critical because it dirierent media or attributes yield similar learning gains and iaalitate achievement oi necessary pertormance criteria, then in a design science or an mstructionai technology, we must always choose the less expensive way to achieve a leamrng goal, We must also form our theories around the under- lying structural ieatures ot the shared properties oi the interchangeable variables and not base theory on the irrelevant suriace teatures. l challenge Robert Kozma and other colleagues in this area to iind evidence, in a well designed study, ot any instance at a medium or media attributes that are rm! replaceable by a diiierent set oi media and attributes to achieve similar learning results ior any given student and learning task. This repioceobiiity test is the key to my argument since ii a treatment can be replaced by another treatment with similar results, the cause oi the results is in some shared (and uncontrolled) properties ot both treatments. or course it is important tor instructional designers to know that there are a variety of treatments that will produce a desired learning goal, However, the utility oi this knowledge is largely economic. The designer can and must choose the less expul- sive and most cognitively etiioent way to rep resent and deliver instniction. it cannot be argued that any given medium or attribute must be present in order ior leaming to occur, only that certain media and attributes are more etiicient tor certain learners, learning goals and tasks. This allows the discussion, and our men- tal set as theorists, to stutt irom media aflnr butes as causal in teaming to media attributes as causal in the coslreflectiveness ot learning. while this may seem a small shitt in the rep- resentation ot the problem, it would have male! consequences tor instructional research, theory and tor design. Cogn ve instructional theory can shitt to a concern with lnskrut‘e