Deception-detection and machine intelligence in practical Turing tests

Deception-detection is the crux of Turing’s experiment to examine machine thinking conveyed through a capacity to respond with sustained and satisfactory answers to unrestricted questions put by a human interrogator. However, in 60 years to the month since the publication of Computing Machinery and Intelligence little agreement exists for a canonical format for Turing’s textual game of imitation, deception and machine intelligence. This research raises from the trapped mine of philosophical claims, counter-claims and rebuttals Turing’s own distinct five minutes question-answer imitation game, which he envisioned practicalised in two different ways: a) A two-participant, interrogator-witness viva voce, b) A three-participant, comparison of a machine with a human both questioned simultaneously by a human interrogator. Using Loebner’s 18th Prize for Artificial Intelligence contest, and Colby et al.’s 1972 transcript analysis paradigm, this research practicalised Turing’s imitation game with over 400 human participants and 13 machines across three original experiments. Results show that, at the current state of technology, a deception rate of 8.33% was achieved by machines in 60 human-machine simultaneous comparison tests. Results also show more than 1 in 3 Reviewers succumbed to hidden interlocutor misidentification after reading transcripts from experiment 2. Deception-detection is essential to uncover the increasing number of malfeasant programmes, such as CyberLover, developed to steal identity and financially defraud users in chatrooms across the Internet. Practicalising Turing’s two tests can assist in understanding natural dialogue and mitigate the risk from cybercrime.

Hidden interlocutor misidentification in practical Turing Tests

Based on insufficient evidence, and inadequate research, Floridi and his students report inaccuracies and draw false conclusions in their Minds and Machines evaluation, which this paper aims to clarify. Acting as invited judges, Floridi et al. participated in nine, of the ninety-six, Turing tests staged in the finals of the 18th Loebner Prize for Artificial Intelligence in October 2008. From the transcripts it appears that they used power over solidarity as an interrogation technique. As a result, they were fooled on several occasions into believing that a machine was a human and that a human was a machine. Worse still, they did not realise their mistake. This resulted in a combined correct identification rate of less than 56%. In their paper they assumed that they had made correct identifications when they in fact had been incorrect.

Some implications of a sample of practical Turing tests

A series of imitation games involving 3-participant (simultaneous comparison of two hidden entities) and 2-participant (direct interrogation of a hidden entity) were conducted at Bletchley Park on the 100th anniversary of Alan Turing’s birth: 23 June 2012. From the ongoing analysis of over 150 games involving (expert and non-expert, males and females, adults and child) judges, machines and hidden humans (foils for the machines), we present six particular conversations that took place between human judges and a hidden entity that produced unexpected results. From this sample we focus on features of Turing’s machine intelligence test that the mathematician/code breaker did not consider in his examination for machine thinking: the subjective nature of attributing intelligence to another mind.

Comment on ``Piecewise potential vorticity inversion: Elementary tests''

Egger (2008) constructs some idealised experiments to test the usefulness of piecewise potential vorticity inversion (PPVI) in the diagnosis of Rossby wave dynamics and baroclinic development. He concludes that, ``PPVI does not help us to understand the dynamics of linear Rossby waves. It provides local tendencies of the streamfunction which are unrelated to the true ones. The same way, the motion of baroclinic waves in shear flow cannot be understood by using PPVI. Moreover, the effect of boundary temperatures as determined by PPVI is unrelated to the flow evolution.'' He goes further in arguing that we should not consider velocities as ``induced'' by PV anomalies defined by carving up the global domain. However, these conclusions partly reflect the limitations of his idealised experiments and the manner in which the PV components were partitioned from one another.

A high-order arbitrarily unstructured finite-volume model of the global atmosphere: tests solving the shallow-water equations

Simulations of the global atmosphere for weather and climate forecasting require fast and accurate solutions and so operational models use high-order finite differences on regular structured grids. This precludes the use of local refinement; techniques allowing local refinement are either expensive (eg. high-order finite element techniques) or have reduced accuracy at changes in resolution (eg. unstructured finite-volume with linear differencing). We present solutions of the shallow-water equations for westerly flow over a mid-latitude mountain from a finite-volume model written using OpenFOAM. A second/third-order accurate differencing scheme is applied on arbitrarily unstructured meshes made up of various shapes and refinement patterns. The results are as accurate as equivalent resolution spectral methods. Using lower order differencing reduces accuracy at a refinement pattern which allows errors from refinement of the mountain to accumulate and reduces the global accuracy over a 15 day simulation. We have therefore introduced a scheme which fits a 2D cubic polynomial approximately on a stencil around each cell. Using this scheme means that refinement of the mountain improves the accuracy after a 15 day simulation. This is a more severe test of local mesh refinement for global simulations than has been presented but a realistic test if these techniques are to be used operationally. These efficient, high-order schemes may make it possible for local mesh refinement to be used by weather and climate forecast models.

The influence of different artificial soil types on the acute toxicity of carbendazim to the earthworm Eisenia fetida in laboratory toxicity tests

Field populations of earthworms have shown a varied response in mortality to the fungicide carbendazim, the toxic reference substance used in agrochemical field trials. The aim of this study was to determine the influence of soil conditions as a potential cause of this variation. Laboratory acute toxicity tests were conducted using a range of artificial soils with varying soil components (organic matter, clay, pH and moisture). Batch adsorption/desorption studies were run to determine the influence of the soil properties on carbendazim behaviour. Adsorption was shown to be correlated with organic matter content and pH and this in turn could be linked to Eisenia fetida mortality, with lower mortality occurring with increased adsorption. Overall while E.fetida mortality did vary significantly between several of the soils the calculated LC50 values in the different soils did not cover a wide range (6.04-16.00 mg kg(-1)), showing that under these laboratory conditions soil components did not greatly influence carbendazim toxicity to E.fietida. (c) 2007 Elsevier Masson SAS. All rights reserved.

Did farming arise from a misapplication of social intelligence?

The origins of farming is the defining event of human history-the one turning point that has resulted in modern humans having a quite different type of lifestyle and cognition to all other animals and past types of humans. With the economic basis provided by farming, human individuals and societies have developed types of material culture that greatly augment powers of memory and computation, extending the human mental capacity far beyond that which the brain alone can provide. Archaeologists have long debated and discussed why people began living in settled communities and became dependent on cultivated plants and animals, which soon evolved into domesticated forms. One of the most intriguing explanations was proposed more than 20 years ago not by an archaeologist but by a psychologist: Nicholas Humphrey suggested that farming arose from the 'misapplication of social intelligence'. I explore this idea in relation to recent discoveries and archaeological interpretations in the Near East, arguing that social intelligence has indeed played a key role in the origin of farming and hence the emergence of the modern world.

A capture–recapture approach for screening using two diagnostic tests with availability of disease status for the test positives only

The article considers screening human populations with two screening tests. If any of the two tests is positive, then full evaluation of the disease status is undertaken; however, if both diagnostic tests are negative, then disease status remains unknown. This procedure leads to a data constellation in which, for each disease status, the 2 × 2 table associated with the two diagnostic tests used in screening has exactly one empty, unknown cell. To estimate the unobserved cell counts, previous approaches assume independence of the two diagnostic tests and use specific models, including the special mixture model of Walter or unconstrained capture–recapture estimates. Often, as is also demonstrated in this article by means of a simple test, the independence of the two screening tests is not supported by the data. Two new estimators are suggested that allow associations of the screening test, although the form of association must be assumed to be homogeneous over disease status. These estimators are modifications of the simple capture–recapture estimator and easy to construct. The estimators are investigated for several screening studies with fully evaluated disease status in which the superior behavior of the new estimators compared to the previous conventional ones can be shown. Finally, the performance of the new estimators is compared with maximum likelihood estimators, which are more difficult to obtain in these models. The results indicate the loss of efficiency as minor.

The effect of consumption volume on profile and liking of oral nutritional supplements of varied sweetness: sequential profiling and boredom tests

Oral nutrition supplements (ONS) are routinely prescribed to those with, or at risk of, malnutrition. Previous research identified poor compliance due to taste and sweetness. This paper investigates taste and hedonic liking of ONS, of varying sweetness and metallic levels, over consumption volume; an important consideration as patients are prescribed large volumes of ONS daily. A sequential descriptive profile was developed to determine the perception of sensory attributes over repeat consumption of ONS. Changes in liking of ONS following repeat consumption were characterised by a boredom test. Certain flavour (metallic taste, soya milk flavour) and mouthfeel (mouthdrying, mouthcoating) attributes built up over increased consumption volume (p 0.002). Hedonic liking data from two cohorts, healthy older volunteers (n = 32, median age 73) and patients (n = 28, median age 85), suggested such build-up was disliked. Efforts made to improve the palatability of ONS must take account of the build up of taste and mouthfeel characteristics over increased consumption volume.

Phyto-activity of biocides used to manipulate herbivory: tests of three pesticides on fourteen plant species

1. We tested three pesticides used for field manipulations of herbivory for direct phytoactive effects on the germination and growth of 14 herbaceous plant species selected to provide a range of life-history strategies and functional groups. 2. We report three companion experiments: (A) Two insecticides, chlorpyrifos (granular soil insecticide) and dimethoate (foliar spray), were applied in fully-factorial combination to pot-germinated individuals of 12 species. (B) The same fully-factorial design was used to test for direct effects on the germination of four herbaceous legumes. (C) The molluscicide, metaldehyde, was tested for direct effects on the germination and growth of six plant species. 3. The insecticides had few significant effects on growth and germination. Dimethoate acted only on growth stimulating Anisantha sterilis, Sonchus asper and Stellaria graminea. In contrast, chlorpyrifos acted on germination increasing the germination of Trifolium dubium and Trifolium pratense. There was also a significant interactive effect of chlorpyrifos and dimethoate on the germination of T pratense. However, all. effects were relatively small in magnitude and explanatory power. The molluscicide had no significant effect on plant germination or growth. 4. The small number and size of direct effects of the pesticides on plant performance is encouraging for the use of these pesticides in manipulative experiments on herbivory, especially for the molluscicide. However, a smatt number of direct (positive) effects of the insecticides on some plant species need to be taken into account when interpreting field manipulations of herbivory with these compounds, and emphasises the importance of conducting tests for direct phyto-active effects. (C) 2004 Elsevier GmbH. All rights reserved.

A capture-recapture approach for screening using two diagnostic tests with availability of disease status for the test positives only

The article considers screening human populations with two screening tests. If any of the two tests is positive, then full evaluation of the disease status is undertaken; however, if both diagnostic tests are negative, then disease status remains unknown. This procedure leads to a data constellation in which, for each disease status, the 2 x 2 table associated with the two diagnostic tests used in screening has exactly one empty, unknown cell. To estimate the unobserved cell counts, previous approaches assume independence of the two diagnostic tests and use specific models, including the special mixture model of Walter or unconstrained capture-recapture estimates. Often, as is also demonstrated in this article by means of a simple test, the independence of the two screening tests is not supported by the data. Two new estimators are suggested that allow associations of the screening test, although the form of association must be assumed to be homogeneous over disease status. These estimators are modifications of the simple capture-recapture estimator and easy to construct. The estimators are investigated for several screening studies with fully evaluated disease status in which the superior behavior of the new estimators compared to the previous conventional ones can be shown. Finally, the performance of the new estimators is compared with maximum likelihood estimators, which are more difficult to obtain in these models. The results indicate the loss of efficiency as minor.