Loading presentation...
Prezi is an interactive zooming presentation

Present Remotely

Send the link below via email or IM

Copy

Present to your audience

Start remote presentation

  • Invited audience members will follow you as you navigate and present
  • People invited to a presentation do not need a Prezi account
  • This link expires 10 minutes after you close the presentation
  • A maximum of 30 users can follow your presentation
  • Learn more about this feature in our knowledge base article

Do you really want to delete this prezi?

Neither you, nor the coeditors you shared it with will be able to recover it again.

DeleteCancel

Make your likes visible on Facebook?

Connect your Facebook account to Prezi and let your likes appear on your timeline.
You can change this under Settings & Account at any time.

No, thanks

Player of Games presentation

No description
by

Alessandro Canossa

on 26 November 2016

Comments (0)

Please log in to add your comment.

Report abuse

Transcript of Player of Games presentation

FALLOUT NV MOD
inclusive designs
Detecting personality/exploring individual differences
through gameplay analysis
Openness to Experience
: Fantasy, Aesthetics, Feelings, Actions, Ideas, Values

Conscientiousness
: Competence, Order, Dutifulness, Achievement Striving, Self-Discipline, Deliberation

Extraversion
: Warmth, Gregariousness, Assertiveness, Activity, Excitement-Seeking, Positive Emotions

Agreeableness
: Trust, Straightforwardness, Altruism, Compliance, Modesty, Tender-mindedness

Neuroticism
: Anxiety, Hostility, Depression, Self-consciousness, Impulsiveness, Vulnerability
Five Factor Model - Traits and Facets
For each location, we performed a partial correlation analysis on each behavioral metric versus each personality trait (n=5) and facet (n=30) score.

Controlling for previous game experience.

Analysis I - Aggregated
B =
f
(P, E)
Behavior: actions that occur in a physical and social environment -micro “turns up corner of lips” and molar “smiles frequently”
From Demographics
to Psychographics

Psychology of Personality
Method
Fallout New Vegas mod (FPS - RPG)
Condensed all gameplay affordances in 45 minutes of play
7 microlocations / situations
Very varied environments / situations, each identified by unique configuration of affordances
Multiple choices for each situations

Very granular instrumentation (telemetry)
41 participants
Personality test (300-item IPIP NEO)
Basic demographics and gameplay expertise

Practical Example
adaptivity
alternative to demographics segmentation
job / performance profiling
(Lewin 1935)
(Mischel 1938, Murray 1938)
Personality: behavioral tendencies over time and across situations (traits predict behavior broadly, facets predict more specific behaviors)
Environment/Situation: not 'noise' but context within which behavior takes place (alpha press "objective properties" and beta press "subjective interpretation")
Limitation:
none of the previous studies examined the role played by facets and situations in determining player behaviors

Traits, not facets

Aggregated, not contextualized
Previous Work
Five Factor Model - Facets and Situations
Traits predict behavior broadly,
facets
predict more specific behaviors
The expression of facets in each
situation
enhances the prediction accuracy while narrowing the range of prediction.
Traits = habitual patterns of behavior, thought, and emotion
Extraversion
(concerned with what is outside the self):
Warmth,
Gregariousness,
Assertiveness,
Activity-Level,
Excitement-Seeking
Positive Emotions.

Assertiveness and Gregariousness = orthogonal factors
Activity Level and Warmth = unrelated facets

Extraversion = not particularly informative
but individual facets score = provide more detail on individual behavior.
Facets = specific and unique aspects of broader personality traits
IntroHouse
The starting location is in a guestroom in the house of the mayor of a small town. Here players awaken from a shock that caused memory loss, furthering a trope often seen in games.


Tracy: tutorial
Mrs. Walker: optional quest
Mr. Walker: main quest, bikers

Socializing with the family, accepting a side quest, melee combat, pick up a melee weapons. There is not much room for exploration beside a couple of rooms that serve no real purpose.

Town:
small rural settlement, anonymous also in its layout, a simple crossroad bisecting the valley north to south and east to west.

David Johnson: main quest.
Buildings: the mayor’s house, the bar and the sheriff’s office.
Three different classes of NPCs generic NPCs that have short dialogues,info rich NPC
quest giver NPC


The sheriff’s office:
A small, deserted and fairly dark environment, offering the chance to pick up a gun and a stack of money. As the main quest can be solved both by paying off the bikers at the hotel or shooting them, this location provides instruments for both alternatives. This gives the player the choice of being either accommodating or aggressive when dealing with the bikers.
The bar:
A spacious environment, several tables, plenty of objects and a cramped back room.
Two generic NPCs
Shotgun hidden in the back.
Large numbers of objects to collect, catering for hoarding behaviors, several areas to sit, and an interactive jukebox. By offering several options to interact both with unanimated objects and characters, it becomes possible to gauge whether players are more oriented towards people or things. The shotgun in the back is only available to players that are willing to explore the space.
Abandoned Farm
In town, Richard Woods an info rich NPC mentions an abandoned farm located just out of town, overrun by giant rats. This is not a formal quest, Richard Woods does not ask players to clear the farm rats, no progress is tracked and no completion is rewarded. Nevertheless, players that speak with Richard and then visit the farm can showcase high level of initiative, attentiveness and conscientiousness.
Jeff’s Shack
Just out of town, on the side of the main road, players can meet Jeff, a generic NPC, hanging out in front of his dilapidated shack. Jeff mentions a cave full of coyotes and provides loose instructions on how to find the cave. As with Richard Woods and the abandoned farm, this is not a formal quest, Jeff does not ask players to clear the cave from the, no progress is tracked and no completion is rewarded. Nevertheless, players that speak with Jeff and then visit the cave can showcase high level of initiative, attentiveness and conscientiousness.

Coyotes’ cave
The cave is fairly difficult to find, unless players received directions from Jeff. The cave is a very dark arena populated by seven hostile coyotes that will attack on sight. Players can decide to stay and fight or leave, providing indication on their attitudes. The cave has no connection to any quest; its existence is limited to gauging exploration and creature interaction.

Abandoned house
An abandoned, dilapidated house, fairly spacious and illuminated but cluttered with objects of no importance and no creatures. There is no reason for players to spend any time here, yet players that are very curious or dutiful might spend a long time trying to uncover something of value.

Silver mine
An easy location to find. As soon as players enter the mine they are approached by Ann, asking for help in locating and liberating her husband, Mark, captured by the bikers that invaded town. This is the beginning of a structured, self-contained side quest. The mine is a fairly large and illuminated space structured in two levels. In the upper mine the bikers have taken residence. Spider is the leader of the pack, if approached he will ask for some silver as ransom to free Mark, alternatively it is possible to engage him and the gang in combat and free Mark after the fight. The lower mine is a fairly large and complex space, where players have to look for a suitcase full with silver, the only means to liberate Mark in a peaceful manner. As soon as Mark is free, he will ask players to report his new condition to Ann, if players chose to do so the quest will be considered complete.

The Hotel:
The hotel is a spacious and brightly illuminated log cabin organized in two floors. Upstairs are eight rooms, where several bikers are resting. The lower floor is organized around the reception desk; on the left we find the billiard room, with several bikers playing pool, while on the right is a bar lounge. Behind the reception are more rooms guarded by bikers, here the sheriff is kept prisoner. Players entering the hotel are confronted with Slater, the gang’s boss, see figure 8. Slater instantly engages aggressively players with a dialogue that dictates how the confrontation should play out: by paying the bikers with money retrieved in the sheriff’s office, it is possible to free the sheriff and liberate the town from the gang peacefully; alternatively it is possible to engage in combat and defeat the gang, freeing both sheriff and town. The last option consists of circumventing Slater, locate the sheriff, provide him with a gun and let him take care of the gang alone or aiding him in that task.

The valley chokepoint:
divide world in two parts: the area where the town is located, together with all other locations, and another small valley containing the Hotel. The town and the hotel are connected by a main road and separated by a chokepoint. The purpose of separating the hotel from the rest of the map is to give the hotel the relevance and focus deserved by the final showdown for all players following the main quest, setting it apart from players that are just wandering aimlessly the town and the locations in its periphery. Players merely exploring should not be faced with heavy opposition and a potential confrontation with a large number of bikers, the risk would be to bias the behavior of players for the rest of the experience.
Conversational Interactions
- six peaceful interaction behaviors:
- interactions with NPCs,
- speech choices making small talk,
- speech choices ending dialogue,
- time spent in dialogue,
- instances of dialogue,
- time spent in the proximity of NPCs.
Narrative Compliance
- behaviors aligned with completing tasks assigned to the players:
- creatures and NPCs attacked that were quest related,
- number of quests accepted or completed,
- speech choices inquiring about quests,
- total time spent on each of the five game quests.
Combat Behaviors
- aggressive interactions and attitudes:
- creatures and NPCs attacked unmotivated,
- left mouse button pressed (attack).
Navigation and Interaction with the world
- nine behaviors:
- number of areas entered,
- interactions with doors,
- speech choices inquiring about the world,
- total distance traveled,
- total head movement,
- interactions with dead creatures and NPCs,
- interactions with containers,
- time spent looking in containers,
- key ‘E’ pressed (activate objects).
Behavioral Metrics

Repeated the analysis, accounting for context by grouping behavioral measurements by game areas

The number of correlations performed, were organized into “cells”.

One cell contained a single personality trait with its facets, in a single location, with all the behaviors of a single class (navigation, conversation or narrative).

The numbers represent # of significant correlations
Analysis II - Context
Type I error is a serious concern.
To rule out spurious correlations we performed a Montecarlo bootstrap as already previously used by Yee et al.
Any cell that had fewer significant correlations than the bootstrap cutoff was considered non-significant as a whole.
+ p<0.10 * p<0.05
Analysis III - Examining the Cells
Overall, the inclusion of facets and situations helped account for behavior that traits alone could not and greatly increased correlation coefficients from previous work
+ p<0.10 * p<0.05 ** p<0.01
We examined correlations between behavioral measures (Navigation, Conversation, Narrative) and traits aggregating the entire game session, without accounting for location.

Only one significant relation between Openness to Experience and Quest Time (rho=-0.35 p<0.05).

Expanding the analysis to include the facets yielded just four more results: Gregariousness vs. Distance Traveled (rho = 0.38), Assertiveness vs. Quest Time (rho = -0.33), and Achievement vs. both Head Movement and Door interaction (rho = 0.34 and 0.41).

Aggregating the data without accounting for game area (context/situation) was insufficient.
Navigation Outside has strongly significant correlations with Activity Levels a facet of Extraversion. It is the location mostly characterized by navigation Behaviors by the whole population
Conversation in the Hotel is strongly correlated with Achievement Striving and Self-discipline: in order to talk with the biker gang leader it's necessary to navigate a tricky dialogue tree to avoid devolving into a gunfight
Cutoff
: In order for a behavior to be subjected to the location-specific analysis, it had to be exhibited by at least 90% of the players for that area.
intro house
outdoor
intro house
outdoor
High Openness
Low Openness
High Openness
Low Openness
Impact of Motivation, Personality and Context on Gameplay Behavior (and why does it matter)
Demographics

age,
gender,
declared hours of play,
level of education,
gaming preferences.
Psychographics:
Self-determination Theory (PENS)
California Adult Q-Set
Attributional Complexity scale
Ego-Control scale
Hostility scale (CMH)
Empathy scale
High-Sensitive Person scale
Multidimensional Perfectionism scale
Need for Cognition scale

Reiss Motivation Profile
NEO-PI-R (Five Factor Model)
Narcissism scale (NPI)
Personal Need for Structure scale
Rational Experiential Inventory (Epstein)
Self-Esteem scale (Rosenberg)
Satisfaction with Life
Self-Consciousness Scale
Self-Disclosure scale


basic psychometric tools
Gameplay behavior
is a function of personality and context,

Gender, education, age, ethnicity have less impact on behavior than personality

Preliminary Study
Correlation with personality (RMP)
Mean and standard deviation of Minecraft players compared to general population

Acceptance -0,6279* 1,364
Curiosity 0,9767** 1,355
Honor 0,5698* 1,164
Idealism 0,6977* 1,311
Independence 0,6977* 1,293
Order -0,4302* 1,306
Saving 0,6860* 1,119
Status -0,9767** 1,208

Pearson correlation between 16 basic life motives and 546 stats from Minecraft
Correlations with demographics
• Age is negatively correlated with the number of jumps: Young people seem to skip through the level.
• Age is negatively correlated with multiplayer joins. Young people join multiplayer games; older people tend to play alone.
• Women are positively correlated with glass and cakes crafted, distance swum, bows depleted.
• Age is positively correlated with the number of deaths, older players die more.
• Gaming preferences show that the more player’s dedication the more it uses and crafts objects made with diamond and less objects made with stone and wood.
• People with strong curiosity tend to use (and reuse) more torches. They also use more stone objects.
• People with strong family reject the use of gold swords; it could be due to its lack of efficiency to protect people.
• Players with strong honor use more rail tracks and craft/use bows consistently.
• Players with strong idealism deal less damage and use fewer cacti.
• People strongly motivated by power place fewer flowers.
• Saving is related with the type of material used, people with strong saving tend to use “cheap materials”: stone, cheap, sand, and iron and place fewer amounts of beds. People with weak saving use diamonds materials.
• Players reporting strong status do not use cactus but tend to use potatoes.
• People reporting strong vengeance tend to quit the game more times, maybe because of the few aggressive affordances provided by the game.

Demographics: quantifiable statistics of a given population in time, used to characterize and segment it.

age,
gender,
declared hours of play,
level of education,
gaming preferences.
92 participants, filling the RMP for 16 life motives, 20 hours minecraft play and demographic questionnaire
Psychographics Correlations
Mean and standard deviation of Minecraft players compared to general population


Curiosity 0,9767** 1,355
Status -0,9766** 1,208
Idealism 0,6977* 1,311
Independence 0,6977* 1,293
Saving 0,6860* 1,119
Acceptance -0,6279* 1,364
Honor 0,5698* 1,164
Order -0,4302* 1,306



Pearson correlation between 16 basic life motives and 546 stats from Minecraft
Demographics Correlations
• People with strong curiosity tend to use (and reuse) more torches. They also use more stone objects.
• People with strong family reject the use of gold swords; it could be due to its lack of efficiency to protect people.
• Players with strong honor use more rail tracks and craft/use bows consistently.
• Players with strong idealism deal less damage and use fewer cacti.
• People strongly motivated by power place fewer flowers.
• Saving is related with the type of material used, people with strong saving tend to use “cheap materials”: stone, cheap, sand, and iron and place fewer amounts of beds. People with weak saving use diamonds materials.
• Players reporting strong status do not use cactus but tend to use potatoes.
• People reporting strong vengeance tend to quit the game more times, maybe because of the few aggressive affordances provided by the game.

• Age is negatively correlated with the number of jumps: Young people seem to skip through the level.
• Age is negatively correlated with multiplayer joins. Young people join multiplayer games; older people tend to play alone.
• Women are positively correlated with glass and cakes crafted, distance swum, bows depleted.
• Age is positively correlated with the number of deaths, older players die more.
games as diagnostic tools to assess valuable and difficult constructs: relisience, grit, etc...
five factor model
raw gameplay metrics
participant's multimodal heatmap
Attack Quest Related
Number of Kills
Attacks Initiator
Weapon of choice
Conversational Interactions
Number of Conversations
Time Spent on Conversations
Time spent around NPCs
Dialogue Choices
Navigation
Number of Areas Visited
Time Spent in each Area
Movement modifier
Interaction with the world
number of pick up/drop
number of doors opened/closed
number of interactions with containers
time spent in inventory/pip boy
Combat behavior
Narrative compliance
Number of Quests accepted
Number of Quests completed
Time spent on Quest
Traits = habitual patterns of behavior, thought, and emotion

Facets = specific and unique aspects of broader personality traits

Extraversion (concerned with what is outside the self):
Warmth,
Gregariousness,
Assertiveness,
Activity-Level,
Excitement-Seeking
Positive Emotions.






Assertiveness and Gregariousness = orthogonal factors

Extraversion = not particularly informative
but individual facets score = provide more detail on individual behavior.
In Your Face(t)
Data analyses were conducted in two phases.

Phase 1 represents an analysis of total-game behavior,

Phase 2 represents a collection of location-based analyses.
Analysis Plan
- make use of facets
- account for context
When interacting with digital games,
people have no mouth and they must scream,
they have no hands and they must pull the trigger, they have no limbs and they must escape.
This is what I do: try to read meaning in mute voices, vacuous gestures and fleeting trails of movement

Openness to Experience
: Fantasy, Aesthetics, Feelings, Actions, Ideas, Values

Conscientiousness
: Competence, Order, Dutifulness, Achievement Striving, Self-Discipline, Deliberation

Extraversion
: Warmth, Gregariousness, Assertiveness, Activity, Excitement-Seeking, Positive Emotions

Agreeableness
: Trust, Straightforwardness, Altruism, Compliance, Modesty, Tender-mindedness

Neuroticism
: Anxiety, Hostility, Depression, Self-consciousness, Impulsiveness, Vulnerability
and the cool promises of adaptivity, profiling, assessment?
the lesson from upside down owl...
Full transcript