Does Spatial or Visual Information in Maps Facilitate Text Recall? Reconsidering the Conjoint Retention Hypothesis 1C Marlynn M. Griffin Daniel H. Robinson The conjoint retention hypothesis (CRH) claims that students recall more text information when they study geographic maps in addition to text than when they study text alone, because the maps are encoded spatially (Kulhavy, Lee, & Caterino, 1985). This claim was recently challenged by Griffin and Robinson (2000), whofound no advantagefor maps overfeature lists infacilitating text recall. In two experiments, we crossed maps and lists with icons and names (c.f., Griffin & Robinson), and employed materials and methodology very similar to those used in previous CRH studies by Kulhavy and colleagues (Kulhavy, Stock, Verdi, Rittschof, and Savenye, 1993; Stock, Kulhavy, Peterson, Hancock, & Verdi, 1995). In addition, we included a concurrent task to measure spatial encoding, as did Griffn and Robinson. No advantages werefoundfor maps over lists in facilitating text recall, nor were maps processed in a more spatial manner than lists. Instead, it appears that the key stimulus featureforfacilitating text recall is mimetic icons (i.e., icons that representfeatures) rather than the spatial characteristics of geographic maps, afinding that supports dual-coding theory (Paivio, 1986), but not the CRH. 0 Numerous studies have shown that provid- ing students a geographic map as an adjunct to text results in better recall of text information that is also referenced in the map than when stu- dents are provided only the text itself (e.g., Abel & Kulhavy, 1986; Kulhavy, Stock, Verdi, Rittschof, & Savenye, 1993; Schwartz & Kulhavy, 1981). Kulhavy and his colleagues (e.g., Kulhavy, Lee et al., 1985; Kulhavy & Stock, 1996) have explained this facilitative advantage of geographic maps using the conjoint retention hypothesis (CRH). According to the CRH, infor- mation that is referenced in a text and an accom- panying map is encoded both verbally and spatially, whereas information found only in text is encoded verbally. The map and text infor- mation are "conjointly retained" offering a dual coding advantage (Paivio, 1986). This linked information (spatial and verbal) provides a retrieval advantage because the spatial repre- sentation can be accessed when retrieval of the verbal representation fails (Kulhavy, Stock, Peterson, Pridemore, & Klein, 1992). Later reconceptualizations (Kulhavy & Stock, 1996; Kulhavy, Woodard, Haygood, & Webb, 1993) of the CRH have attempted to further explain the facilitative advantage of maps, using three major lines of reasoning. First, verbal descriptions are far less adequate for represent- ing the structural and inferential relations con- tained in even a simple map. Second, maps are processed differently by the memorial system than are verbal descriptions, with descriptions processed sequentially while "depictions allow unordered and simultaneous access to visual details" (p.128; see also, Das, Kirby, & Jarman, 1979). Third, Tippett (1992) noted that visual ETR&D, Vol. 53. No. 1, 2005. pp. 23-36 ISSN 1042-1629 23