Duffrin et al. (2001) used pawpaw fruit puree as a fat-reducing agent in muffins, compared to muffins made with applesauce and fat. These authors found that muffins made with pawpaw puree (≈91% fat replacement) were equally acceptable to the other muffins for all characteristics, with the exception of appearance. The study concluded that pawpaw fruit puree had the potential as a carbohydrate-based fat reducing agent in muffins and suggested that it had the potential to be used as a fat reducing agent in other types of baked goods. Continued investigation of pawpaw fruit puree as a fat-reducing agent in baked goods develops an area of potential use for this product and provides consumers with another product option that could be used to lower dietary fat intake.

Hydrogenated vegetable oil is used in a plain shortened cake food formulation (Penfield and Campbell, 1990) and contributes to the sensory qualities of flavor, tenderness, and color (Swanson and Munsayac, 1999) in the finished cake product. Although these characteristics are desired in the finished cake product, the trans fatty acid intake associated with the use of hydrogenated vegetable oil is not a desirable component of a heart healthy diet. Trans fatty acids have been found to be a contributor to heart disease by increasing blood cholesterol levels (Jacobs and Murtaugh, 2000; Weisburger, 1997).

Many fruit such as apples (Malus ×sylvestris var. domestica), prunes (Prunus domestica), (Swanson and Munsayac, 1999) and pawpaws (Duffrin et al., 2001) have been used as a fat-reducing agent in baked goods. Using a carbohydrate-based fat-reducing agent in a plain shortened cake would produce a food formulation lower in trans fatty acids, total fat, and calories. The objective of this study was to investigate the effects of substituting varying levels of hydrogenated vegetable oil with pawpaw fruit puree on the sensory properties of a plain shortened cake.

ing varying levels of hydrogenated vegetable oil with pawpaw fruit puree on the sensory properties of a plain shortened cake.

Cakes were prepared using a plain shortened cake recipe modified from Penfield and Campbell (1990). The conventional method of mixing was used to prepare the cake batter. The cakes were prepared by replacing about 25%, 50%, and 75% of the fat with pawpaw fruit puree (Integration Acres, Albany, Ohio) and compared to a control using 100% vegetable shortening. With the exception of the pawpaw, all ingredients were obtained from a local supermarket. The cake formulations are listed in Table 1.

All dry ingredients and eggs were weighed using an electronic scale (Acculab Inc., Newton, Pa.) accurate to 0.1 g. The milk was measured in a 50-mL graduated cylinder and the vanilla was measured in a 10-mL graduated cylinder. The flour, baking powder, and salt were sifted together in a medium-mixing bowl. The vanilla was added to the milk. A handheld electric mixer (model 62520; Hamilton Beach, Picton, Ont., Canada) was used at a speed of 3 for all mixing. The shortening was creamed for 1 min and the sugar was added and mixed for an additional min. Egg was then added to the creamed mixture and mixed for an additional 2 min. One-third of the flour mixture and one-third of the milk was added and mixed for 45 s, this was repeated twice. The mixing times were kept consistent for each sample preparation. The batter was poured into aluminum anodized round cake pans [5.1 cm (2 inches) deep and 20.3 cm (8 inches) diameter (Marpol Inc., Rock Falls, Ill.)] that were lightly coated with vegetable oil spray. The samples were baked at 176.7 °C (350 °F) for 35 min and were removed from the pans after cooling for 15 min.

35 min and were removed from the pans after cooling for 15 min.

Cake samples [2.5 cm (1 inch) square] were placed on individually coded, square sampling dishes. The four samples were placed on a tray and simultaneously presented to the panelists. Samples were tasted in random order. Participants were provided with tepid (room temperature) water to cleanse the palate between samples. A scorecard with a hedonic scale was used to rate each cake sample for color, flavor, texture, tenderness, aftertaste, and overall acceptability. The scale was composed of eight choices ranging from 1 = like extremely, to 8 = dislike extremely.

A sample of untrained, volunteer, adult consumer panelists (n = 105) were invited from the university campus to participate in sensory evaluation of the cake samples. Panelists were given information, both verbally and in writing, about the expectations of their participation in the study. After signing a participation consent form, panelists were seated at individual testing stations with standard fluorescent light and given a scorecard with written instructions asking them to rate their cake samples. Descriptive statistics, analysis of variance (ANOVA), and post-hoc analysis (Tukey's honestly significant difference) were completed using the Statistical Package for the Social Sciences (SPSS, Chicago); P < 0.05 was considered significant.

Mean scores suggested that panelists liked all cake samples to some degree (slightly to extremely) (Table 2). The 25% sample received more favorable mean ratings in the categories of flavor and texture, although panelists detected no differences between the control and 25% fat replacement sample for all sensory attributes. ANOVA of