(AGENPARL) – mar 07 febbraio 2023 Si chiama “Brave New Word” il progetto di ricerca, coordinato da Marco Marelli, professore del dipartimento di Psicologia, che svilupperà un modello in grado di predire idee ed intuizioni stimolate quando si imparano vocaboli sconosciuti. Tanti i possibili campi di applicazione, dall’intelligenza artificiale all’educazione, dal marketing alla riabilitazione linguistica
Comunicato stampa
Milano-Bicocca, dall’ERC 1,4 milioni di euro per scoprire come l’apprendimento di nuove parole porti a pensare in maniera differente
Si chiama “Brave New Word” il progetto di ricerca, coordinato da Marco Marelli, professore del dipartimento di Psicologia, che svilupperà un modello in grado di predire idee ed intuizioni stimolate quando si imparano vocaboli sconosciuti. Tanti i possibili campi di applicazione, dall’intelligenza artificiale all’educazione, dal marketing alla riabilitazione linguistica
Milano, 31 gennaio 2023 – Sviluppare un modello in grado di predire i significati, le idee e le intuizioni che vengono stimolati nel momento in cui si apprendono nuove parole e come questi si relazionano con le conoscenze e credenze pregresse.
Gli ERC – Consolidator Grant sono finanziamenti destinati dal Consiglio Europeo della Ricerca a ricercatori con un’esperienza pluriennale maturata dopo il conseguimento del dottorato che hanno l’obiettivo di consolidare la propria indipendenza nella ricerca creando o rafforzando – se già costituito – un proprio gruppo di lavoro.
Negli ultimi dieci anni, l’Università di Milano-Bicocca ha ricevuto finanziamenti per 13 progetti ERC, 12 con il programma della Commissione Europea “Horizon 2020” (2014-2020, di cui 5 Starting Grant, 5 Consolidator Grant, 1 Proof of Concept, 1 Synergy Grant) e 1 con il nuovo programma “Horizon Europe” (2021-2027), il Consolidator Grant ottenuto” da “Brave New Word”.
«Questi risultati – afferma il prorettore alla Ricerca Guido Cavaletti – confortati dall’ultimo risultato positivo ottenuto dal professore Marco Marelli, sono la dimostrazione tangibile del valore della ricerca dell’Università di Milano-Bicocca e della tensione dei ricercatori dell’Ateneo verso la dimensione europea. Per un’Università giovane e multidisciplinare come la nostra questa è l’unica strategia in grado di garantire da un lato un ulteriore sviluppo, e dall’altro di restituire alla società civile risultati di reale rilevanza, anche grazie allo sviluppo di efficaci reti internazionali».
Il titolo del progetto trae spunto dal capolavoro dello scrittore e filosofo Aldous Huxley (“Brave new world”, in italiano “il mondo nuovo”) e fa riferimento al «coraggio della nuova parola nello stimolare nuove idee», introduce la ricerca Marco Marelli. «L’apprendimento di nuove parole è un processo che, a tutti gli effetti, non ha mai fine – spiega il professore del dipartimento di Psicologia –. Per quanto la maggior parte del vocabolario della nostra lingua madre, ovvero le parole che conosciamo, venga acquisito in età dello sviluppo, persino da adulti impariamo una nuova parola ogni due giorni. Raramente ce ne rendiamo conto: le nuove parole vengono integrate nel vocabolario preesistente in maniera automatica, senza sforzo esplicito, o persino consapevolezza, da parte nostra».
«Con le nuove parole – prosegue Marelli – acquisiamo anche nuovi concetti ed idee: apprendere nuove parole arricchisce il nostro mondo interno, e ci può portare a pensare in modo differente. Il progetto “Brave New Word” mira a studiare questo fenomeno combinando psicologia sperimentale, linguistica computazionale, e neuroscienze cognitive. L’obiettivo è lo sviluppo di un modello in grado di predire, su base matematica, le idee e i significati che vengono stimolati dall’esperienza di nuove parole, e come questi si relazionano con le nostre conoscenze e credenze pregresse».
Il modello si baserà su tre indizi linguistici che tipicamente accompagnano l’esperienza di nuove parole:
– informazioni contestuali, “intorno” alla parola (le frasi in cui le nuove parole vengono utilizzate);
– la struttura della parola stessa. «Come nel famoso caso di “petaloso” – spiega Marelli –: per quanto discutessimo se fosse una vera parola o meno, quando questo termine venne divulgato, tutti avevamo una intuizione chiara del suo significato, basata sugli elementi “petalo” e “oso” di cui era composta»;
– associazioni tra suoni e significati derivati dall’esperienza linguistica: «tipicamente – osserva il professore – una parola come “sundaliera”, dal suono più dolce, ci evocherà significati maggiormente positivi di “turtotioro”, dal suono più duro».
Alla definizione del modello contribuirà una serie di esperimenti dal vivo ed online, sia sulla lingua italiana, che sulla lingua inglese, che coinvolgerà centinaia di individui. «I diversi studi analizzeranno le intuizioni suscitate da nuove parole, e valuteranno quanto queste siano in linea con le predizioni quantitative fornite dal modello, permettendoci di affinarlo. Le intuizioni verranno raccolte attraverso diversi metodi di misurazione, quali associazioni libere, tempi di reazione, mouse-tracking, e attivazione cerebrale (tramite elettroencefalogramma e risonanza magnetica funzionale)».
«Il lavoro proposto si caratterizza come ricerca di base – conclude Marco Marelli – ma una volta sviluppato, il modello potrà trovare diversi naturali ambiti di applicazione, dal contributo allo sviluppo di una Intelligenza artificiale sempre più in linea con la mente umana, alla definizione di programmi di educazione e riabilitazione linguistica, fino al marketing e alla comunicazione».

Testo Allegato:
3997496-102327931 January 202302000031 January 2023October – December 2022Contractual wages and salariesThe index of wages according to the national collective bargaining agreement measures the evolution of wages and salaries (per employee or per hour) determined by contractual provisions set by collective agreements; the indices are calculated with reference to the fixed employment structure of the base period (December 2015). At the end of December 2022 the coverage rate (share of national collective agreements in force for the wage setting aspects) concerns 50.7% of employees – about 6.2 million – and 51.2% of the total amount of wages. In December 2022 the hourly and the per employee indices increased by 0.1 from the previous month. Compared with December 2021 the hourly and the per employee indices increased by 1.5%.On average in 2022, both indices grew by 1.1% compared to 2021. CHART 1. INDEX OF WAGES PER HOUR ACCORDING TO COLLECTIVE AGREEMENTS January 2015 – December 2022 (Index, December 2015=100)CHART 2. WAGES PER HOUR ACCORDING TO COLLECTIVE AGREEMENTSJanuary 2015-December 2022 (Index, December 2015=100), month on previous month, month on same month a year ago percentage changes and annual average changesCHART 3. WAGES PER HOUR ACCORDING TO COLLECTIVE AGREEMENTS, BY ECONOMIC SECTOR January 2015-December 2022, month on same month a year ago percentage changes (Index, December 2015=100)TABLE 1. WAGES per hour and per employee ACCORDING TO COLLECTIVE AGREEMENT October – December 2022, indices, month on previous month and on same month a year ago percentage changes (Index, December 2015=100)Index(December2015=100)Month on previous month percentage changesMonth on same month a year agopercentage changesOctNovDecOct 22Sep 22Nov 22Oct 22Dec 22Nov 22Oct 22Oct 21Nov 22Nov 21Dec 22Dec 2120222021Wages according to collective agreements per hour106.1106.3106.4+ 0.2+0.2+0.1+1.3+1.3+1.5+1.1Wages according to collective agreements per employee106.1106.4106.5+0.1+0.3+0.1+1.2+1.5+1.5+1.1CHART 4. BARGAINING TENSION INDICATORS. EMPLOYEES AWAITING RENEWAL, TOTAL AND PRIVATE SECTORJanuary 2015-December 2022, percentage valuesCHART 5. BARGAINING TENSION INDICATORS. AVERAGE NUMBER OF MONTHS PASSED FROM THE EXPIRATION DATE, TOTAL AND PRIVATE SECTORJanuary 2015-December 2022Contractual length of work: hours of work to be worked, according to the collective agreements, by full time employees, net of those paid but not worked, for holidays, work permits (annual work reduction, abolished public holidays, study, union assembly).Employee: Employees (full time equivalent), excluding apprentices and managers, whose compensations are ruled by the national collective bargaining agreement in sectors.Employees awaiting renewal: the share of employees covered by collective agreements expired.National Collective bargaining agreement: written contract negotiated through employers’ association and trade unions that regulates the terms and conditions of employees at work. The index numbers of the contractual wages per employee provide a measure of the variations of remunerations due, over the course of a year, to the employed worker on the basis of the contents of the national labour contract and the legislation in force. The indices of the contractual remunerations per employees are based on the equivalent work units (EWU), namely full and part-time employees reported to work units, without considering any detraction for any absences, nor other reasons that justify a reduction of the remuneration provided by contract.The index of the hourly contractual remunerations: is obtained as the ratio between each elementary index of the remunerations and the corresponding index of the contractual length of work, measures monthly the variations of the contractual remunerations to pay to employees for each hour of work contractually performed.Wages according to collective agreement: contractual wages calculated as one twelfth of annual wages and salaries on the basis of the national collective labour agreements. The accounted components are: base salary, cost of living allowance, periodic seniority increases, shift work allowance and other general allowances (when relevant in the sector), additional months of salary and other payments regularly paid in the year term.Introduction and regulatory frameworkThe contractual wages and salaries and working hours data are taken from national collective bargaining agreements, or from laws and regulations that govern the topic. It is necessary, however, to remember that the index of the contractual wages and salaries has purely national characteristics, and it’s not included among those subject to European Regulation. In addition to being the most timely indicator of the evolution of wages and salaries, it assumes particular importance because it is the basis of numerous legislative provisions (and not) for the adjustment of wages and salaries, pension and fee amounts to specific categories of employees and services.Index reference baseThe series of the indices use December 2015 as the reference base. The choice to refer the base to a month (rather than a year), is linked to the indicator characteristics, which does not present seasonal components because it refers to annual remuneration. Year 2015 was chosen, analogously to that established for the other short-term indicators by the European Regulation on short-term statistics. The new series of indices of contractual remunerations are calculated starting from January 2015; therefore, these indices replace, from January 2015 to December 2018, the previously indices published in base 2010. Starting from January 2019, the new series of indices are appropriate to produce the legal effects that the laws in force reconnect to specific indicators calculated by Istat.Observation field: information sources and acquisition methodsTo determinate the monthly index, the survey follows the evolution ofthegross remunerations and work hours for all the employment categories involved, except apprentices and managers.In total, there are 73 national collective bargaining agreements monitored by the survey, while there were 78 for the base 2010 (Table 1). The difference of the number of agreements is due to the merging of the previously distinct contracts (the mobility – contractual area of railway activities contract that includes also the contracting services companies from the railways; the maritime transport contract that includes the ship’s crew section and amministrative staff; the air transport contract that join together pilots, flight crew and ground staff).National agreements monitored in the private sector are 58 while totally national contracts are about 280. For the agriculture and construction sectors, provincial contracts are also considered, whose clauses give rise to a remuneration that integrates national remuneration. In the public sector, the agreements already monitored have been confirmed. Researchers and technologists are also included in the research institute sector. This employees represent 96.3%of totally public sector. The contractual remuneration treatment regards 2,855 specific professional profiles (Table 1), characterised by qualification and level of employment. This professional profiles have a different remuneration fixed by the collective contracts.The total profiles are 2,940, of these 1,418 are defined by national contracts (95 relative to public managers who are not included in the monthly index calculation) and 1,522 are fixed by territorial agreements (1,123 for agriculture and 399 for costruction).For each economic activity sector, the national collective bargaining agreements considered are the most representative for the number of employees respect to the other contrats of the same sector. As consequence the leader contracts dynamic is attribuited to the group of employees belonging to contracts represented by them. TABLE 1. contractual wages. year 2015. Contracts, employees, distribution of the remuneration and number of employees. MAIN CONTRACT GROUPINGSCCNLEmployeesRemuneration (%)Number of profilesBlue collarsWhite collarsTotalAgriculture2326,9131.909911321,123Industry244,179,65534.19623316939Private services325,032,30239.53171295466Total private sector589,538,30275.621.7857432.528Public Administration152,825,14224.38-327327Total monthly index7312,364,012100.001,7851,0702,855Public Administration managers under contract9157,375–4444Public Administration managers not under contract590,809–4141Total8712,612,196–1,1552,940The reference employment structureThe reference population is represented by employees: white collars, blue collars and middle managers. The estimate of the number of jobs to which attribute the remuneration dynamic, is derived from several archives. The agricultural sectorFor the agricultural sector, the number of white collars was calculated directly from the Emens archive, while for the blue collars, the data used was taken from the quarterly declaration module (DMAG module) that the farms must sent to INPS (National Social Security Institution) for the survey of the employed labour force. This information has also allowed to determinate the number of temporary and permanent workers. The temporary employees, infact, has wages tables different from permanent workers. The extra-agricultural private sectorThe estimate of the number of jobs full-time for the extra-agricultural private sector is calculated from RACLI (Annual register on earnings, working hours and labour cost for persons and enterprises), based on UniEmens models. This information is integrated with the economic activity presents in ASIA (archive of active company).The analysis have generally been conducted at group level of Ateco, but for particular situations, they have also extended to the class level.For the construction sector, information taken from the administrative archives of the Building Security Institutions was used to produce indices at provincial level.Public AdministrationIn the public sector the number of employees referring to the “Conto Annuale” of the State General Accounting Department on the date of 31 December 2015.Particular Ateco groupsAs in the previous base, some group Ateco are not represented. The exclusion occurred when more than 50% of employees are regulated by agreeements not included in the group considered. For particular Ateco, under the 50%, the class level has been analysed to decide to include them or not (Table 2). TABLE 2. Group and Class Ateco excluded By surveyYears 2010 and 2015GROUP/CLASS OF ECONOMIC ACTIVITYPresence20102015031Fishingnot presentNOT032AquacultureNOTYES051Mining of hard coalpresentNOT107Manufacture of bakery and farinaceous productsNOTYES1071Manufacture of bread; manufacture of fresh pastry goods and cakesYESNOT1072Manufacture of rusks and biscuits; manufacture of preserved pastry goods and cakesYESNOT1073Manufacture of macaroni, noodles, couscous and similar farinaceous productsYESNOT182Manufacture of cocoa, chocolate and sugar confectioneryNOTYES321Manufacture of jewellery, bijouterie and related articlesNOTNOT325Manufacture of medical and dental instruments and suppliesNOTwhite collars477Retail sale of other goods in specialised storesNOTYES4771Retail sale of clothing in specialised storesYESNOT4772Retail sale of footwear and leather goods in specialised stores YESNOT4773Dispensing chemist in specialised storesNOTNOT4774Retail sale of medical and orthopaedic goods in specialised storesYESNOT4775Retail sale of cosmetic and toilet articles in specialised storesYESNOT4776Retail sale of flowers, plants, seeds, fertilisers, pet animals and pet food in specialised storesYESNOT4777Retail sale of watches and jewellery in specialised storesYESNOT4778Other retail sale of new goods in specialised storesYESNOT4779Retail sale of second-hand goods in storesYESNOT591Motion picture, video and television programme activitiesNOTYES592Sound recording and music publishing activitiesNOTwhite collars643Trusts, funds and similar financial entitiesYESNOT653Pension fundingYESNOT662Activities auxiliary to insurance and pension fundingYESNOT683Real estate activities on a fee or contract basisNOTYES742Photographic activitiesNOTwhite collars781Activities of employment placement agenciesYESwhite collars782Temporary employment agency activitiesNOTNOT900Performing artsNOTblue collars931Operation of sports facilitiesNOTNOT941Activities of business and employers membership organisationsYESNOT942Activities of trade unionsYESNOT949Activities of other membership organisationsYESNOT960Other personal service activitiesNOTNOT9601Washing and (dry-)cleaning of textile and fur productsYESYES9602Hairdressing and other beauty treatmentNOTNOT9603Funeral and related activitiesNOTNOT9604Physical well-being activitiesNOTNOT9609Other personal service activities n.e.c.YESYES970Activities of households as employers of domestic personnelNOTNOT982Undifferentiated service-producing activities of private households for own useNOTNOT990Activities of extraterritorial organisations and bodiesNOTNOTContractual elements considered in the monthly indicesFor each of the categories of employees provided by national collective bargaining agreements observed by the survey, the annual per capita remuneration is calculated on the basis of the values in force.The survey considers the remuneration elementshaving a general and continuous feature, including the additional monthly payments and other amounts paid only in some periods of the year; occasional bonuses are excluded, overtime work and emoluments established by decentralised agreements and arrears and lump sum payments. The remuneration elements are presented in Table 3.The following elements are common to all the sectors: base pay, contingency allowance and distinct element of the remuneration; with regard to the other elements, the remuneration structure varies from sector to sector.The annual duration of the work is made up of a number of hours due by contact in the year, net of hours paid, but not worked, for vacations, holidays and paid leaves established by the agreements. The elements considered for the calculation of the annual duration are:1. weekly contractual hours;2. vacations;3. mid-week holidays and recovery of cancelled holidays;4. hours of assembly;5. hours of study;6. hours reduced from the annual amount, used as paid leave.TABLE 3. ELEMENTS OF REMUNERATION CONSIDERED IN THE INDICATORS1Base pay (minimum wage)2Cost of living allowance (the amounts are fixed at the levels achieved in November 1991; there are many sectors in which the bargaining has provided the pooling of the base pay)3Seniority-linked bonuses (included in the calculation of the indices with reference, generally, to an average seniority fixed at 8 years)4Shift work remuneration (in sectors in which the size of the phenomena is noteworthy)5Remuneration for particular activity (connected to the inconvenience or danger, in sectors in which it is possible to estimate an adequate amount with reference to particular “contractual figures”)6Any bonuses7Distinct element of pay (EDR)8Advances and allowances of contractual vacation9Ad personam cheques (generally determined at the time of “reclassification” operations) provided by a new national contract regulationAMonthly wages (sum of items from 1 to 9)10Share of paid holidays11Share of the 13th month of salary12Share of the 14th month salary and of any other additional salary months13Share of any other establishments of annual characterBTotal on monthly base of the accruals of the annual amounts divided into months (sum of items from 10 to 13)A+BWages divided into months ( sum of items from 1 to 13)Characteristics of the indicatorsThe monthly indicator of “contractual wages” is determined from national collective bargaining agreements. The indicator refers to a concept of “price” of the performance of employed work, different from the total wages because it refers to a constant group of workers, and characterised by a fixed composition by qualification, class of employee and average seniority. The index of contractual remuneration is not affected by changes in employee composition. , by the amount of hours actually worked and by the payment of specific non-continuous remunerations (arrears, lump sum, etc.).The index of the contractual wages per employee measures the variations of remuneration to be paid to the employed workers, in a year, on the basis of the national agreeement and the legislation in force. . The remuneration is calculated as one twelfth of the remuneration due over the entire year. The indices of the contractual remunerations per employees are based on the equivalent work units (EWU), namely full time and part-time employees reported to work units, without considering any detraction for absencesor other reductions of hour not remunerated.The index of the contractual length of work measures the variations of the quantity of work that is requested by contract.. For eachclass of employee,the index is based on the number of hours required by contract over the entire year, less the hours paid but not worked for vacation, holidays and otherpaid leaves established by the agreements. The index of the hourly contractual remunerations is obtained as the ratio between elementary index of the remunerations and the corresponding index of the lenght of work.This index measures monthly the variations of the contractual remunerations to pay to employees for each hour of work contractually performed.The “elementary index” (refered to a level of contractual classification, separately for blue collars and white collars), is calculated as a ratio between the remuneration of a specific month and the remuneration of the reference period base (December 2015). n ijt=n rjtn rjowhere n is a specific agreement, j is a level of classification of contract n, n rjt is the remuneration at t time and n rjo is period base remuneration.The elementary indices are aggregate to obtain average index for class level (blue collar, white collera and total), for agreements and for sector until the general index (total economy).Lapseyres index is used to aggregate the elementary indices. n It=j=1Jn ijt* n Rj0j=1Jn Rj0where n Rj0 is the production between n rjo and n ejo (employees reffered to december 2015). The index projections The press release presents a projection of the indices of wages for the six months following the reference month. The index projections are based on contractual increases fixed by agreements. The accurancy depends on the share of employees with contracts in full force.Contractual tension indicators.The survey estimates three contractual tension indicators.share of employees with agreements not in force;number of months following the expire date of contract for only employees awaiting renewal;number of months following the expire date of contract for all the employees.The weighting systemFor each sector the contractual wages and salaries indices are summarised through a weighting system that assigns each aggregate a weight equal to the incidence of the relative remuneration respect to the total sector. The remuneration derives from the product between the number of employees for each sector and the average contractual remuneration.In Table 4, the weighting system used for the construction of the indices on the 2015 base is compared with the system used for 2010 base.TABLE 4. INDICES OF THE CONTRACTUAL REMUNERATIONS WEIGHTING STRUCTURE AND NUMBER OF EMPLOYEES FOR THE CONTRACT SECTORS OBSERVED. Comparison between the weighting structures of the year 2010 and 2015.Year 2010Year 2015Employees Weight on the wages amount (a) EmployeesWeight on the wages amount (a)AGRICULTURE347,6111.80326,9131.90INDUSTRY 4,616,50333.944,179,65534.19Mining and quarrying15,5030.1717,3420.22Food and beverages 252,9881.97252,4022.17Textiles, clothing and leather processing378,2332.37364,2502.53Wood, paper and print360,0322.50315,9682.43Energy and petroleum25,1240.2922,0870.28Chemicals204,4401.62208,3971.84Rubber, plastic and processing of non-metallic minerals331,1592.33297,8642.37Metalworking industry2,191,51715.882,063,12316.83Electrical energy and gas80,8950.7768,8100.73Water and waste removal services137,0361.18134,0321.22Building639,5764.86435,3803.57PRIVATE SERVICES (c)5,008,94437.255,032,30239.53Commerce1,947,28214.131,902,36414.80Private Pharmacies–46,1340.40Transports, postal services and connected activities837,9136.67825,9156.96Restaurants and hotels634,5214.04604,4024.07Information and communications services37,9840.4237,1140.43Telecommunications104,6760.77108,4390.88Credit and insurance415,2264.78399,2674.88Other private services1,031,3426.431,108,6677.11TOTAL PRIVATE SECTOR9,973,05872.999,538,87075.62PUBLIC ADMINISTRATION ACTIVITIES2,915,22427.012,825,14224.38Collective bargaining sectors2,437,96721.942,361,26419.74Of which Ministries166,5571.37145,5351.20Regional and local authorities513,4323.92449,8893.39National Health Service557,6495.03528,7884.38School1,026,27210.041,062,5759.14Police Forces314,0023.44298,1763.07Military – Defense129,8481.35131,4741.28Fire-fighting activities33,4070.2834,2280.29Total economy12,888,282100.0012,364,012100.00 (a) Incidence of the contractual wages on the contracts by branch and sector of economic activity on the total economy.TimelinessThe contractual wages indices are disseminated at the end of the month following the reference month.RevisionsThe published data are not subject to revision.IssueThe indices are published monthly on the HYPERLINK “http://dati.istat.it/?lang=en”Istat data warehouse and quarterly with a HYPERLINK “http://www.istat.it/en/archive/contractual+wages”press release.The wages and salaries indices are calculated according: a) contracts and contract groups; b) economic activity on the basis of the Ateco 2007 classification,In the quarterly press release the contractual wages indices are published by contracts and contract groups.In the datawarehouse I.stat (http://www.dati.istat.it), the indices are available monthly by contract and by economic activity (Nace Rev. 2).