Authors: Shawn M. Crowley-Johnson and Diedre J. Ribbens
Johnson Institute of Pie Science
Holiday time brings forth assertions of a strong preference in the source vegetable for the traditional “pumpkin” pie. Qualities such as flavor and texture are cited as the rationale, but rarely is a head-to-head comparison performed. In this study, three types of source vegetables were baked into pies using identical preparation methods. A blind taste test was conducted with five volunteers, who were each asked to identify the source vegetable. None of the volunteers correctly identified the source vegetable for any of the pies. These results indicate it may be the manner of preparation or recipe used, and not the source vegetable, that gives rise to assertions of preference.
The holiday traditon of concluding a Thanksgiving meal with a pumpkin pie began in the nineteenth century [Smith, 2005]. The pie typically consists of roasted and puréed pumpkin, sweetened into a custard with milk and eggs, and seasoned with a blend of spices collectively known as “pumpkin spice.” The custard is baked into a pie shell, cooled, and served with whipped cream.
While many adhere to the tradition of using pumpkin as the source vegetable for their pies, other source vegetables may be substituted and prepared in a similar manner. Sweet potatoes are a common alternative to pumpkin, also dating back to the nineteenth century [Oliver, 1999]. More recently, butternut squash has been suggested as an alternative source vegetable, citing its availability, texture, and flavor as reasons why it is “superior” to pumpkin [Clark, 2012].
The question of which source vegetable makes the best pie has often been the subject of debate, sometimes even among members of the same family. Assertions of preference for one source vegetable over another made by individuals can result in family arguments, insulting the chef or host of the meal, and rivalry among friends and loved ones. Adherents to one camp over another cite flavor, texture, and color as the main reasons for developing their initial preference.
In this study, we asked whether source vegetable preference was due to prior knowledge of the source vegetable used in the pie. Using a blind taste test method, we conducted an experiment to ascertain whether volunteer taste testers could correctly identify the source vegetable in a pie without being told. We found that preparation methods, and not source vegetables, may be responsible for developing preferences among pie eaters.
Materials and Methods
All materials were obtained from a standard grocery store setting.
The pie crust recipe used was graciously dontated by Virginia “Jini” Johnson, founder and matriarch of the Johnson Institute of Pie Science.
The crust recipe is as follows:
Break an egg into a 1 cup measure and beat with a fork. Add 2 tablespoons of white vinegar. Set the liquid aside.
Into the bowl of an electric stand mixer add 5 cups of all-purpose flour, 2 cups of plain vegetable shortening, and 2 teaspoons of salt. Mix well.
Pour the liquid over the flour mixture and mix until just combined. Separate into 5 or 6 balls of dough and wrap in plastic wrap. Allow to rest in the refrigerator for 30 minutes or more. Can keep in the fridge 2-3 days or in the freezer, double wrapped, for up to 6 months.
Three types of source vegetables were prepared using identical methods.
The pie pumpkins and butternut squash were sliced lengthwise and scooped to remove their seeds, then placed cut side down on a baking sheet. The sweet potatoes were rinsed and wrapped in aluminum foil. The source vegetables were then roasted in the oven at 350F for 45-60 min or until they were soft.
The vegetable flesh was then a scooped out of the skins and puréed in food processor. The food processor was cleaned between vegetables to prevent cross contamination. No water was added during the purée process.
Three separate batches of pie filling were prepared, each using a different source vegetable. The same recipe was used to prepare all three pies. This recipe was also generously donated by Jini Johnson.
1.5 cups purée
3/4 c sugar
1/2 t salt
1/2 t ginger
1 t cinnamon
1/4 t nutmeg
3 slightly beaten eggs
1 c milk
1/4 c half-and-half
1 can evaporated milk (6 oz.)
Combine purée, sugar, salt, and spices. Add eggs and milks; blend. Pour into prepared 9″ pie crust. Bake in 450F oven for 10 min, then reduce oven temperature to 325F and bake for 45 min or until a knife inserted in the center comes out clean.
Blind Taste Test
Five volunteers were recruited as taste testers. The testers were not present at the time of pie preparation and were only allowed to see the completed pies. Testers had a mix of declared source vegetable preferences.
The authors succeeded in baking three pies using identical methods as described above. The pies had similar physical appearances, distinguished from each other primarily by different crust decorations as well as slight variations in color (Figure 1). Pie decorations were “stars,” “leaves,” and “dots.” A fourth pie was created as a control using a mix of pie filling from all three source vegetables (leftover filling).The five volunteer taste testers were allowed to inspect the pies and discuss their guesses before tasting. All five tasters correctly identified the mixed control pie prior to tasting due to its smaller size.
The taste testers were allowed to select their own slices of pie and add whipped cream as desired. All testers chose to add whipped cream. Tasting and discussion of source vegetable identification occurred. All five taste testers aligned on which source vegetable they thought was used for which pie, though they were not obligated to generate a consensus.
Taste testers commented that the “stars” pie and the “dots” pie had similarities in flavor and color, while the “leaves” pie had a slightly different flavor and darker color. See Figures 2-4 for close up images of pies.
All taste testers guessed the “stars” pie used butternut squash as the source vegetable, the “dots” pie used pumpkin as the source vegetable, and the “leaves” pie used sweet potato as the source vegetable. They cited the similarities in color and flavor between the “stars” and “dots” pies as the main reason for making this determination.
After their guesses were registered, the true identities of the source vegetables were revealed:
- “Stars” pie used pumpkin
- “Dots” pie used sweet potato
- “Leaves” pie used butternut squash
None of the participants correctly identified the source vegetable for any of the three pies.
Despite previously declared preferences for a particular source vegetable, none of the five taste testers could correctly identify which pie contained their “favorite.” Additionally, all five testers failed to correctly identify the source vegetable for a single pie in the blind taste test. Taken together, these results suggest that the source vegetable used in a pie likely does not give rise to assertions of preference among pie eaters.
Blind taste test method historically used to disprove the idea of preference in food
The idea that an individual has a favorite or preferred taste for one similar-tasting food over another has led to many rivalries. Using a blind taste test to disprove the idea of such preferences is common. Famously, PepsiCo launched a marketing campaign in 1975 called “The Pepsi Challenge,” which used a simple blind taste test to “prove” to people who preferred Coca-cola that they could not tell the difference between Pepsi and Coke, or in some cases actually preferred Pepsi [Wikipedia, 2016].
Removing the predisposed ideas of whether they will like or dislike a food by disguising the food or blinding the taste tester may help alleviate the confounding influence of any long-held beliefs that one food is preferred over another. Additionally, when a taste tester knows they are participating in a study, they may be more inclined to pay attention to small differences in flavor or texture, leading to a more critical analysis of whether they truely like or dislike what they are eating.
In or study, both of these factors may have contributed to the test subjects mis-identifying their favorite source vegetable.
Recipe, not source vegetable, may have greater influence on preference in pies
One of the carefully controlled variables in this study was to use a single recipe for the pies, substituting only the source vegetable used in the purée. Many fine pie recipes exist, and family traditon may mean that a different recipe is used with different source vegetables. Furthermore, recipes can be designed to highlight or bring out the nuanced flavor differences among source vegetables. While we did not test recipe variation in this study, future studies will be aimed at determining the effect recipe choice has on pie and source vegetable preference.
Thank you to Jillian Johnson for making strong assertions of source vegetable preference, prompting the design of this study.
Diedre Ribbens conceptualized the study and Shawn Johnson carried out the pie making and taste tester selection.
Thank you to the volunteer taste test for sacrificing their time and tastebuds in the name of research.