Monday, 25 November 2013

Domestic Violence Protection Order Pilot Scheme

Today sees the launch of the findings of the evaluation of the Domestic Violence Protection Order pilot scheme. The DVPO is a new civil order that has been introduced in Manchester, Wiltshire and West Mercia. Under the terms of the order, a perpetrator of domestic violence can be excluded from the home for up to 28 days. It is used in situations where police would previously have not taken further action (beyond arrest, caution or bail) and where criminal charges are not being brought.

The evaluation was led by Professor Liz Kelly of London Metropolitan University and Professor Joanna R Adler, who is the director of Forensic Psychological Services at Middlesex University. Dr Miranda Horvath was the project manager and Dr Mark Coulson was the quantitative research lead. Key findings are that the DVPO does seem to reduce repeat call-outs to the police, so seems to be associated with a reduction in re-victimisation. This is particularly the case in more chronic cases, where the police had previously attended three or more times. The full report is available here

Today also sees the Home Secretary’s written statement to Parliament regarding the Domestic Violence Disclosures Scheme (Clare’s law) and the DVPO. Both Orders are to be rolled out nationally from March, 2014.

Friday, 15 November 2013

Visiting speaker: Steve Phelps

Title: Longitudinal reciprocity in Chimpanzee Allogrooming

Date: 21 November 2013
Time: 12 noon
Venue: Hatchcroft (HG09)

Many models of social network formation implicitly assume that network properties are static in steady-state. In contrast, actual social networks are highly dynamic: allegiances and collaborations expire and may or may not be renewed at a later date. Moreover, empirical studies show that human social networks are dynamic at the individual level but static at the global level: individuals' degree rankings change considerably over time, whereas network level metrics such as network diameter and clustering coefficient are relatively stable. There have been some attempts to explain these properties of empirical social networks using agent-based models in which agents play social dilemma games with their immediate neighbours, but can also manipulate their network connections to strategic advantage. However, such models cannot straightforwardly account for reciprocal behaviour based on reputation scores ("indirect reciprocity"), which is known to play an important role in many economic interactions. In order to account for indirect reciprocity, we model the network in a bottom-up fashion: the network emerges from the low-level interactions between agents. By so doing we are able to simultaneously account for the effect of both direct reciprocity (e.g. "tit-for-tat") as well as indirect reciprocity (helping strangers) in order to increase one's reputation). We test the implications of our model against a longitudinal dataset of Chimpanzee grooming interactions in order to determine which types of reciprocity, if any, best explain the data. We discuss the importance of the temporal and micro-properties of the data in analysing reciprocity: in particular determining the length of window over which direct reciprocity occurs, and the importance of network-motifs in detecting patterns of indirect reciprocity.

Steve Phelps

University of Essex

Tuesday, 5 November 2013

Departmental RKE bidding deadline extended

Due to mysterious email complications that are beyond my ken it appears some staff did not receive the recent bidding email.

Please note that the deadline for all internal RKE bids has now been extended to:

5pm on Tuesday 19 November 2013

Please email them directly to me.  Sonia has copies of the proforma for all who need it.

Tom Dickins

Saturday, 2 November 2013

Happiness and well-being within cities

A story from The Guardian yesterday should be of interest to a number of folk in our department, and further afield.  It can be found at this link:

The secrets of the world's happiest cities

Tom Dickins

Visiting speaker: Eddy J. Davelaar

Title: Stopping decisions in memory recall

Date: 7 November 2013
Time: 4pm
Venue: Hatchcroft HG03

Cognitive scientific research on memory has provided much insights in our ability to store and retrieve events. However, only recently researchers have focused their attention in our abilities to terminate memory retrieval, despite the ubiquity of the stopping decisions. I will summarise our research progamme into our decisions to terminate memory search. I will highlight its lawful nature as measured with the Exit Latency and show through computational modelling what is needed in order to observe the data. A rational approach to stopping decisions is tested in a design with monetary rewards. Finally, important connections between animal foraging behaviour and memory stopping decisions are drawn and discussed.