Spin the Wheel

Learnability

Pros

• Intuitive betting interface
  • Spinning wheel and presence of coins suggests gambling game.
• Weighted randomness is evident, promoting strategic betting
  • Spinning wheel simulates randomness, while unequal sections of choice wheel suggest weighted probability distribution.
• Affordability for manipulating sophisticated widgets
  • Pile of coins provides affordability to move them onto desired slice on choice wheel. Instructional arrow improves visibility.
  • Handle on side of wheel provides affordability for spinning. Instructional arrow improves visibility.

Cons

• Breakdown of tasks not immediately evident
  • Not obvious there are multiple stages in bidding process (location, price, cuisine), so a first time user may have to transition back and forth between screens when they realize they don't have enough coins to bet in each criteria round.

Efficiency

Cons

• Lots of widgets to manipulate
  • Requires more user interaction and time (dragging coins to desired section on wheel, pressing buttons to move back and forth between rounds, spinning the wheel to make decisions)
• Lots of rounds
  • 3 rounds of betting on each criteria (individual)
  • 3 spins for deciding on each criteria (group)
  • 1 round of betting on final restaurant (individual)
  • 1 spin for deciding on final restaurant (group)
• Lots of turns
  • each player is passed the phone twice 

Safety

• Pros

• Doesn't allow users to type bet amount, and system disables coin dragging when total coins = 0. Therefore, it is impossible to bet more coins than are in the bank. 
• Users can move coins into and out from choice slice on wheel if they change their mind. choices aren’t committed until "next person" arrow button is pressed. 
• Next and Previous arrow buttons allow user to navigate between screens in betting stages (especially important for first time users since they will most likely no know there are multiple criteria to bid on)Cons |

Auction Sale

Slot Machine

Learnability

• Pros

• Only contains standard smart phone widgets, which is consistent to other apps and very intuitive, straightforward.
• People can teach each other because only one person uses it at a time.
• It's obvious that there are different criteria to bid on.

• Cons

• It's not obvious that selection is based on weighted probability distribution.

Efficiency

• Pros

• The bidding processes on different criteria are on one single screen, instead of 3 separate screens as in "Spin the Wheel" and "Auction Sale".
• The spinner used for selecting location, price, type and points is faster to manipulate than putting coins on the wheel or waiting for others' auction decision.

• Cons

• Users have to take time to figure out the allocation of points to make sure that they sum up to 100.

Safety

• Cons

• There is only one screen for different criteria (as opposed to going back and forth between screens), which makes it more likely to screw up everything because the user can't change choices after pressing "Done".

Scenario

Alice, Bob, and Carol haven’t seen each other for a while, and decided that they should meet up tonight and have dinner together. None of them had any concrete suggestions for a restaurant, and all three friends are rather indecisive, so they decided to meet at 77 Mass Ave and then decide where to go. Alice would prefer to go somewhere nearby, but is otherwise indifferent, whereas Bob would prefer to go across the river. Carol doesn’t care where they go, but she’s strongly in the mood for Chinese food, and would prefer that it’s not too expensive (but not too cheap either). Alice, Bob, and Carol need to agree on a specific restaurant that they can all be reasonably happy with. Also, they would prefer to come to an agreement sooner than later, given that they’re standing around outside at the moment, and have some means for narrowing down their options and getting specific suggestions.

Alice, given that she feels strongly about location, suggests going to Central Square, and Bob suggests going to the Back Bay. Carol mentions that she really wants Chinese food tonight, and would prefer that it be moderately priced. Given that information, someone points out that the Prudential Center P. F. Chang’s fits several of their criteria (Back Bay, moderately priced, Chinese food). Even though it doesn’t fit all of their preferences (which it can’t, because Alice and Bob have mutually incompatible preferences for location), it does reasonably accommodate most of them, so the three friends agree that it’s a good choice.

Designs

All designs model the decision-making processing through an interactive game which involves an element of randomness. Spin the Wheel and Slot Machine allows players to place strategic bets on their preferences, whereas Auction Sale allows players to engage in competitive bidding to reach a final restaurant decision. Each design differs in their degrees of user interaction, group collaboration, and completion. To learn more about how each interface works, please peruse the detailed descriptions below along with their corresponding figures.

Design 1: Spin the Wheel

This interface models the decision-making process as a gambling game, where players make bets on different criteria (location, price, cuisine) involved in making the final restaurant decision. The gambling metaphor is represented by a spinning wheel, reminiscent of roulette and other games that involve an element of randomness (e.g., Twister). This game can be broken down into 4 distinct stages, each involving a different number of rounds.

Stage 1: Selection of Criteria Choices (group)

See Figures 2, 3

The group is required to select choices for location (Figures 2,3) and cuisine (not drawn, but similar to Figures 2,3). Figure 2 displays choices for location as equal sections of the wheel, as opposed to the conventional list. This is used for consistency, as the wheel is the centerpiece of the game. 

Stage 2: Bet on Criteria Choices (individual)

See Figures 4, 5, 6 

Each player places bets on each of 3 criteria (location, price, cuisine) by dragging coins from the personal pile on the bottom right-hand corner of the screen to a desired section of the wheel in the center of the screen. Each player is allotted 100 coins for this step, to be distributed (in whatever proportion they choose) amongst the 3 criteria. For example, if a player values location over cuisine type over price, he/she may place 50 coins on location, 30 coins on cuisine, and 20 coins on price.

Stage 3: Decide on Criteria Choices (group)

See Figures 7 (shown for location, extrapolate for price and cuisine), 8

The criteria decision is made by spinning the wheel at the center. This is done 3 times, once for each criteria. The final criteria decision is determined based on a weighted probability distribution, which depends on how many coins were placed on each choice. Before the wheel is spun, the sections of the choice wheel are resized to reflect the probability distribution of deciding each criteria. This weighted probability is based on the coin distribution collected from each player in Stage 2. For example, in Figure 7, 40% of all coins used (among all players) in Stage 2 were placed on Central, whereas only 10% of total coins were placed on Kendall. 

Stage 4: Vote on Restaurant Choices  (individual) -> Output Final Decision (group)

See Figure 9, 10

Each player votes (as opposed to bet in previous rounds) on their favorite restaurant (Figure 9), and the wheel will decide the final restaurant choice (Figure 10), based on weighted probability like in Stage 3. 

Design 2: Auction Sale

This interface models the decision-making process as an auction sale, where participants bid for the opportunity to choose one criteria of the decision-making process (location, cuisine, price, restaurant). Participants are given an allocation of credits to spend during four rounds of bidding. This game can be broken down into 2 distinct stages, with Stage 2 involving multiple rounds for bidding on each criteria.

Stage 1: Login (individual)

See Figures 1 - 5

Each user signs in (Figure 1), and one participant creates a new event and invites the others (Figures 2-5).

Step 2: Bidding on Criteria (individual)

See Figures 6 - 11

(Before bidding begins, choices for each criteria (location, price, cuisine) are automatically generated by the application. Location will be generated based on a sampling of places that are near to the group's geographic position. Prices ($, $$, $$$, $$$$) are default choices, and cuisine types will be randomly generated.)

The decision-making process involves 4 rounds of bidding that work like a normal auction sale (Figures 7-11). Each participant is presented with the highest bid, and his/her own highest bid so far (Figures 8-9). If a participant is not in the lead, he/she may increase his/her bid to one more than the previous highest bid. After a bid has not been made for a certain length of time (~10 seconds), the auction ends, and the winner of that round may choose one aspect of the decision (Figure 10). Others are then informed of the choice (Figure 11). The auction process happens four times. The first 3 rounds determine the location, cuisine, and price range of the final choice. Based on those results, the interface presents the participants with several restaurant choices, and a fourth round of bidding determines the final result.

Design 3: Slot Machine

This interface models the decision-making process as a slot machine game. Like in Spin the Wheel, each player makes bets on different criteria (location, price, cuisine) involved in making the final restaurant decision. The gambling metaphor is represented by a slot machine instead of a spinning wheel. While not so conceptually different from Spin the Wheel, Slot Machine is significantly different in terms of user experience. Most notably, less user interaction is required. Users can bet on the 3 criteria (location, price, cuisine) all in one screen, reducing the amount of time to complete the decision task. 

Stage 1: Selection of Criteria Choices (group)

See Figures 1 - 4

The group is required to select choices for location, price, and cuisine. Each criteria is editable. Pressing "EDIT" (Figure 1) will take the user to a list of current choices for that category (Figure 2). To add a choice, the "+" button is pressed, which will take the user to another screen (Figure 3) that presents a variety of ways to populate the chosen criteria category. For example, choosing "Nearby" (Figure 3) will display a list of nearby locations that the user can check (Figure 4). The user can also search for locations using zipcode and keywords (e.g. Central, MIT, etc) (Figure 3). This will also take the user to a list of corresponding locations that look like Figure 4. A similar process is carried out for price (although this category has less choices - $, $$, $$$, $$$$) and cuisine.  "Done" (Figure 1) is pressed when the group is finished selecting all criteria choices.

Stage 2: Assigning Points to Criteria Choices (individual)

See Figure 5 

This is where the metaphor departs. In this step, we use a slot machine interface for betting on different choices. Like in Spin the Wheel, users assign points to their desired criteria choice. When this is done, the "Decide" button is pressed. The point assignments are saved, and the phone is passed to the next individual in the group. 

Stage 3: Decide on Criteria Choices (group)

See Figure 6

The criteria decision is made by pressing the "Decide It!" button, which will spin the slot machine. The slot machine decides on the final combination of location, price, and cuisine based on a weighted probability of individual choices (just as described in Spin the Wheel). Though randomness is evident, the weighted probability scheme is more hidden to the end user than in "Spin the Wheel", where sections of the choice wheel morphed to represented their probability weighting. 

Stage 4: Vote on Restaurant Choices  (individual) -> Output Final Decision (group)

See Figure 7, 8, 9

The app generates 3 choices for the group to vote on based on the final criteria decisions made in step 3 (Figure 7). Figure 7 allows users to find out more details about each restaurant choice. The "Round 2" button takes the game into the final round, where individuals must vote on their desired restaurant choice (Figure 8). The final restaurant choice is determined by the slot machine based on the weighted probability scheme mentioned in Stage 3 and in Spin the Wheel (Figure 9).

Design Analysis