Shared intention & Joint action

Question 1

Joint/social attention: joint intentionality

Answer 1

Tomasello: progress of...
individual intentionality
-> joint intentionality
-> collective intentionality
... is what makes humans unique. 

Joint/shared intentionality produces collaborative behaviour to achieve a common goal (Tomasello et al, 05).

People with different perspectives work differently toward the same outcome.

Question 2

Joint/social attention: social eyes (gazes)

Answer 2

Eyes are the windows to the soul - is intention revealed?

Eye gaze, head and body orientation - tell our social intentions (Nummenmaa & Calder, 2009).

Eyes can tell others what you’re interested in.

Question 3

Joint/social attention: eye gaze as a social signal

Answer 3

Shows overt attention: indicates info important to an individual.

• also important for others to investigate.
• found in dogs, dolphins, primates etc.

Shows covert attention: not necessarily readable.

• human infants gaze following starts early (Farroni et al, 2000; Tomasello, 2001).
• shift/check attention = 9-12 months.
• follow others attention = 11-14 months.
• ability to direct attention = 13-14 months.

Human interaction important to determine progression.

Question 4

Gaze leading: gaze following

Answer 4

Sometimes to check whether we are being attended to.

Want to find out if people are following us.

If person doesn’t follow - may not have shared interests.
- impacts future engagements.

Question 5

Gaze leading: attending the gazing face

Answer 5

Tomasello (2005): joint attention is collaborative.
- not one follows the other.
Don’t just follow eye gaze but also lead.

Edwards et al (2015): do we observe the fazing face?

• gaze followed condition vs. averted.
• look at cross, then gaze initiated by another stimulus in centre.
• one face follows gaze, one averts from stimulus.
• then random letter appears on either face.

Findings:

• quicker RTs when letter on face follow ppts gaze.
• more attention paid - shared interests.

Question 6

Gaze leading: to be watched or not?

Answer 6

Dual functions of eye gaze - gather info and deliver message.

Primates and humans use direct eye contact to exert superiority (De Waal, 1989).

• gaze can also tell us whose in charge/hierarchy.
• people of different status use different gaze patterns.
• cultural variations - inappropriate looking somewhere or at someone for too long.

Goebel et al, (2015): ppts watched video clips of higher vs. lower ranked targets.

• told filmed video would be stored (non-social) vs. watched by targets (social).
• look less to eyes of higher ranked targets in social condition - avoid challenging superiority?

Question 7

Joint action: explicit interaction

Answer 7

Don’t just engage in interplays with eye gaze… also in actions.

Need to perform complimentary actions rather than imitating another person - to achieve a common goal (Sebanz et al 2006).

Question 8

Joint action: compared to social facilitation

Answer 8

Fundamentally different from SF (Zajonc, 1965).

• people behave differently when being observed = audience effect.
• or when carrying out identical tasks = co-action effect.

Typically better performance for simple tasks but worse for complex/new tasks.

Joint action = complimentary (not identical) tasks.

Question 9

Joint action: simon effect

Answer 9

Two choice (L/R hand) responses according to colours, but the finger image can point to either direction (Simon, 1969).

• stimulus-response conflict can occur: longer RTs for incompatible condition (responding hand differs from pointing direction) - Knoblich & Sebanz, 2006).

Eg. either arm up = respond with left hand // down = right response.

• conflict between visual-spatial info when right hand up (respond left).
• less difficult when one task is removed - no difference in RTs.

Question 10

Joint action: social simon effect

Answer 10

Two responses taken by two people (Sebanz et al, 2003/2005):
- same compatibility effect found when a pair performed the two parts of the simon task.
Compared to individual cognition (Knoblich, 2006).
- RTs significantly longer for incompatible tasks in the joint condition.
- RTs a only a little slower for incompatible tasks in individual condition.

Question 11

Joint action: flanker task

Answer 11

Slower responses to target when info is conflicting/incongruent (Eriksen & Eriksen, 1974).

Question 12

Joint action: social flanker task - Atmaca et al (2011)

Answer 12

Joint condition: left person press when middle letter = H or K.
- right person when = C or S.

Individual task: press when = H or K. Do nothing when = C or S.

Stronger flanker effect in joint condition.

Question 13

Joint action: task co-representation

Answer 13

Whole task processed (represented) by one person - even if only performing part of the task.

Suggested that this person represents the partaking on by the other person as well = task co-representation or shared task representation.

“Basic interpersonal processes put to service by more advanced functions that support the type of intentionality required to engage in joint action, cultural learning and communication”. (Knoblich & Sebanz, 2008).
- not the same as social facilitation (improvement in a task when working with others - better on simple tasks, worse on complex tasks).

Question 14

Is joint action social?: relationships

Answer 14

This effect is stronger when two persons have a positive relationship (Hommel et al, 2009).
- RTs longer for positive relationship - greater difference between spatial non-correspondence and spatial correspondence.

Question 15

Is joint action social?: believing is enough

Answer 15

Social Simon effect found when ppt believed there was another co-actor (fake) - Sebanz et al, 2005.
- social flanker effect also - Atmaca et al, 2011.

Co-representation only occurs if believed to be human, not computer - Tsai et al, 2008.

But effect present with wooden hand when person takes perspective of Pinocchio (Muller et al, 2011).
- greater differences in RTs in Pinocchio condition, incompatible task.

Question 16

Is joint action social?: intention is key

Answer 16

Social flanker only occurred when co-acting person was performing intentionally (Atmaca et al, 2011).

In unintentional condition, action of the co-actor was controlled by a machine.

Question 17

Is joint action social?: not always clear

Answer 17

Social Simon also found in the presence of non-living objects (Dolk et al, 2013) - wooden cat figures.

Question 18

Implicit interpersonal processes: interpersonal processes

Answer 18

= “the interplay of cognitive, motivational and behavioural activities in social interaction.” (Snyder & Stukas, 1999).

Interaction can either be explicit or implicit.

Joint action = explicit in that each co-acting person is performing a task within integrates task set or common goal.

Question 19

Implicit interpersonal processes: going implicit

Answer 19

Human beings shaped evolutionarily - social interactions are so important for survival; social behaviour may end up implicit.
- don’t have to voluntarily put effort and pay attention for social processes to occur.

Even without active engagement, humans still take in social info surrounding us into mental representation of environment.
- embedded in environment automatically and subconsciously.

E.g. “saw a friend who briefly spoke about interior design. Looking in food and wine section of a book store, attention is draw to interior design book even though you did not care about it at all”.

• info relevant to others influences us and not in a helpful way.
• How? (He et al, 2014).

Question 20

Implicit interpersonal processes: intrapersonal memory guidance

Answer 20

Working memory content can guide visual attention: goes to memorised items first (monkeys - Chelazzi et al, 1993; human study - Soto et al, 2008).

Faster responses to targets next to memorised item (valid condition) - than targets far away (invalid condition).
- lose keys but remember where phone is so quicker to find phone and realise keys are next to it.

Two people perform tasks similar but separate (cannot be integrated together as a single task).

No incentive for the participants to form co-representation.

Adapted the memory guidance paradigm for a dyad (He et al, 2011; 2014).

Preview (to be memorised by oneself/co-actor/nobody):

• visual search (two attention conditions: valid vs. invalid).
• memory test.

Question 21

Implicit interpersonal processes: interpersonal memory guidance

Answer 21

Three effects to check (He et al, 2014):

1. Intrapersonal = memory guidance by ones own stimuli.
2. Interpersonal = memory guidance by the other person’s stimuli.
3. Baseline - guidance by the presence of stimuli (no memory involved; usually no guidance effect).

Question 22

Implicit interpersonal processes: group membership

Answer 22

This effect only found in strangers (He et al, 2011).

Social effect - group membership = key factor in RT.

Question 23

Implicit interpersonal processes: collectivism

Answer 23

Individual-collectivism: the degree to which ppts view themselves as members of a social group or an individual (Triandis, 1995).

Collectivist: take on group values and norms as a member in that group.

Individualistic: view themselves as loosely connected to other people and choose to operate autonomously.

Positively correlated - collectivism vs interpersonal memory guidance (He et al, 2014).

Question 24

Implicit interpersonal processes: implicit task co-representation

Answer 24

Co-actors task is implicitly co-represented without any subjective intentions or consciousness.

Not all tasks are co-represented, because the memory performance per se is comparable across groups (He et al, 2011).
- only the effect of memory on attention is interpersonally affected.

Question 25

Shared attention: social = sharing

Answer 25

All social interaction are realised via shared reality.

- i.e. the presence of all parties at the moment.

Question 26

Shared attention: shared attention

Answer 26

Doesn’t require active interpersonal interaction (eg. gaze following).
- similar to implicit interpersonal processes.

A person’s psychological status in which attention is shared according to his/her belief.
- “a person may believe they are sharing attention to a movie screen with another, when in fact their partner is asleep” (Shteynberg, 2015).

Relationship plus synchrony (Shteynberg, 2015). 
- conditions of emergence: 
> relationally close to others OR
> synchronously co-attending. 
- leads to - shared attention state: 
> " we are attending to X".
- leads to - impact of shared attention:
> greater cognitive resources to X. 
- leads to - empirical findings:
> better memory. 
> stronger motivation. 
> more extreme judgements. 
> higher effective intensity. 
> greater behaviour and learning.

Question 27

Self prioritisation: self

Answer 27

In social interactions we share intentions and attention and sometimes work toward a common goal.

• how do I know what I need to do to achieve the group goal?
• part to individually play in a shared project - separate parts to achieve common goal come together as a whole.

Egocentric framework.

Question 28

Self prioritisation: self is a special framework

Answer 28

Show processing advantage to info relevant to ourselves.
- more crucial to our survival and has higher ecological saliency.

Self reference effect (SRE): people encode info differently, depending on relevance to self.

• achieve better recall rate for self-relevant info (Conway & Dewhurst, 1995).
• “friendly” - better remembered after being asked if it describes you vs. Donald Trump.

More self related bias examples:

• memory for objects owned by oneself better vs. others (Conway et al, 1996).
• one also has better face recognition to own face than other faces (Sui et al, 2003).
• people tend to choose positive traits to describe themselves vs. others (Klein et al, 1989).

Arguments against SRE: self related stimuli might be more familiar or rewarding than alternatives.

Question 29

Self prioritisation: is self really special?

Answer 29

Need to address criticisms by using more neutral stimuli (Humphreys & Sui, 2015).

It is possible that self is so deeply embedded into cognitive processing that low-level cognition is also biased for self relevant information?

Question 30

Self prioritisation: social association paradigm & self prioritisation effect

Answer 30

Association learning between social labels (you, friend, stranger with a name) and geometric shapes respectively (circle, square, triangle) - Sui et al, 2012.
- ppts needed to report whether the shape-label was correct or not.

Self prioritisation effect: found faster and more accurate responses for self associations (Sui et al, 2012).

• shapes equally familiar and counterbalanced.
• labels matched for usage and length frequency.
• shapes don’t carry social meanings - self advantage effect.

Perceptual sensitivity measured with shapes contrast reduced - more difficult to be seen (Sui et al, 2012).

Perceptual sensitivity to self associations not affected, friend associations harder to perceive.

Question 31

Self prioritisation: automatic effect?

Answer 31

Effect from an automatic process, once association is established?
- or controlled by top-down expectation?

Question 32

Self prioritisation: automatic

Answer 32

Means self-prioritisation is hard to be biased against.

• cannot be explained by familiarity or expectancy.
• suggested self has high social saliency and is automatic.

Question 33

Self prioritisation: is self treated as a reward?

Answer 33

Reward association similar result to self association (Sui et al, 2012).
- try to overcome self bias using reward.

Question 34

Self prioritisation: self bias resists reward

Answer 34

E.g. stranger (£8), friend (£2), self (£0) - Humphreys and Sui (2015).

• RTs as fast as those high reward stranger vs. high reward friend - both faster than intermediate friend reward.
• self always faster to respond even when reward against.

Not the same for friend/strangers - if the value from a stranger is strong enough then rewards could trump friendship.

Question 35

Self prioritisation: baby bias

Answer 35

Can still behave selflessly.

- baby association shows advantage over the mother’s self (Sui & Humphreys, unpublished).

Question 36

Self prioritisation: neural correlates (Sui et al, 2013)

Answer 36

Self-related processing:

• ventral medial prefrontal cortex (VMPFC): reward processing.
• left posterior superior temporal sulcus (LPSTS): salient, contextual processing.

Stranger-related processing:
- dorsal top-down attention network: new task demands processing.